{ "schema": "yizuo-agent-materials-plaza/catalog@2", "categoryEnumVersion": 1, "categories": [ "pack", "skill", "mcp", "facade", "official-data" ], "generatedAt": "2026-11-02", "count": 285, "counts": { "packs": 75, "skills": 197, "mcpServers": 13, "tools": 649 }, "materials": [ { "id": "bQ-NlOBloZegGfkiczktDA", "slug": "pack-bq-nloblozeggfkiczktda", "category": "pack", "marketCategory": "facet", "tags": [ "官方功能集", "星研", "学院", "教程" ], "canonical": { "brief": "Xingyan Academy's first release, two lessons: 1) 'Getting Started with Yizuo' — your first deliverable: document library, pack install, a real tool call, trace check, sharing; 2) 'Intro to Wanxing and Facet Collaboration' — assemble skills and MCP into a personal workflow, deployed as a callable ser" }, "human": { "zh-CN": { "name": "星研·学院", "summary": "星研·学院首辑两课：①《壹座上手》——从零完成第一个可交付任务：文档库、功能集安装、真实工具调用、trace 自检、分享交付；②《萬星与谦面协作入门》——把技能与 MCP 组装成个人工作流，部署为可被程序调用的服务（A2A、访客页、控制台）。安装后你的助手兼任助教，按课逐步带你动手完成。" }, "en": { "name": "Xingyan Academy", "summary": "Xingyan Academy's first release, two lessons: 1) 'Getting Started with Yizuo' — your first deliverable: document library, pack install, a real tool call, trace check, sharing; 2) 'Intro to Wanxing and Facet Collaboration' — assemble skills and MCP into a personal workflow, deployed as a callable service (A2A, visitor page, console). Your assistant doubles as a teaching assistant and walks you through each lesson." }, "es": { "name": "Academia Xingyan", "summary": "Primera entrega de Xingyan, dos lecciones: 1) «Primeros pasos con Yizuo» — tu primer entregable: biblioteca, instalación de packs, una llamada real de herramienta, autochequeo de trace, compartir; 2) «Introducción a la colaboración Wanxing y Facet» — arma habilidades y MCP en un flujo personal, desplegado como servicio invocable (A2A, página visitante, consola). Tu asistente te guía lección a lección." }, "fr": { "name": "Académie Xingyan", "summary": "Première livraison de Xingyan, deux leçons : 1) « Premiers pas avec Yizuo » — votre premier livrable : bibliothèque, installation de packs, un vrai appel d'outil, autocontrôle de trace, partage ; 2) « Initiation à la collaboration Wanxing et Facet » — assemblez compétences et MCP en un flux personnel, déployé comme service appelable (A2A, page visiteur, console). Votre assistant fait office d'assistant de cours, leçon par leçon." }, "ja": { "name": "Xingyan アカデミー", "summary": "星研・学院の第一弾は 2 課：①『壹座はじめの一歩』——ゼロから最初の成果物を完成させる：ドキュメント庫、機能セット導入、実際のツール呼び出し、trace セルフチェック、共有まで；②『萬星と Facet の協作入門』——スキルと MCP を個人ワークフローに組み上げ、プログラムから呼び出せるサービスとしてデプロイ（A2A、訪問者ページ、コンソール）。導入後はあなたのアシスタントが助教を兼ね、課ごとに手を動かしながら案内します。" } }, "whenToUse": "Browse under \"facet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "xingyan-academy-yizuo", "description": "星研·学院《壹座上手：从零完成第一个可交付任务》（use 层）伴学技能：用壹座完成第一个可交付任务：把自有资料喂进文档库、装上一个行业能力包、让它带真实工具调用产出结论，最后交出一条匿名可读的分享链接。全程在浏览器里完成，不需要装任何软件、不需要命令行。 当用户询问本课怎么开始、某一步怎么做、卡住怎么办时使用。" }, { "name": "xingyan-academy-wanxing-facet", "description": "星研·学院《萬星与谦面协作入门：搭起你的第一个个人工作流》（build 层）伴学技能：把壹座里装好的技能与 MCP 组装成属于你的个人工作流（自建预设），把它部署到萬星成为一个可被程序调用的服务，并用一条 A2A 请求、一个访客对话页和控制台把这项服务管起来。 当用户询问本课怎么开始、某一步怎么做、卡住怎么办时使用。" } ], "mcpServers": [] } }, { "id": "_FN7pti9TH5JMg3YnQut0w", "slug": "pack-fn7pti9th5jmg3ynqut0w", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "量化", "策略", "回测", "wire" ], "canonical": { "brief": "Wire Strategy Lab, a four-skill set: strategy design (weight semantics and universe validation), backtest interpretation (five requirements and three prohibitions), risk-control parameters (meaning / magnitude / limitations), and paper-run supervision (run events / weight snapshots / data gaps). Div" }, "human": { "zh-CN": { "name": "柏讯 · 策略工坊", "summary": "柏讯策略工坊四件套：策略设计（权重语义与 universe 验证）、回测解读（五条要求与三条禁止项）、风控参数（含义/量级/局限）、模拟盘督导（运行事件/权重快照/数据缺口）。与柏讯模拟交易套件的分工：那边是操作面——开账户、下单、撤单、订单历史，由对话操作技能承接；本套件是思考面——想清楚权重怎么定、回测结果怎么读、风控参数意味着什么，一切成交仍走模拟交易服务。paper-only：无真实资金；回测与模拟盘不预示未来。" }, "en": { "name": "Wire · Strategy Lab", "summary": "Wire Strategy Lab, a four-skill set: strategy design (weight semantics and universe validation), backtest interpretation (five requirements and three prohibitions), risk-control parameters (meaning / magnitude / limitations), and paper-run supervision (run events / weight snapshots / data gaps). Division of labor with the Wire Paper Trading suite: that one is the operations plane — open accounts, place orders, cancel orders, order history, handled by conversational operation skills; this suite is the thinking plane — getting clear on how weights should be set, how to read backtest results, what risk parameters mean; every fill still goes through the paper trading service. Paper-only: no real money; backtests and paper runs do not foreshadow the future." }, "es": { "name": "Wire · Laboratorio de estrategias", "summary": "Cuatro skills de Wire · Laboratorio de estrategias: diseño de estrategia (semántica de pesos y validación del universe), interpretación de backtests (cinco exigencias y tres prohibiciones), parámetros de control de riesgo (significado / magnitud / límites) y supervisión del paper run (eventos de ejecución / instantáneas de pesos / lagunas de datos). Reparto con la suite de paper trading de Wire: aquella es el plano operativo — abrir cuentas, enviar órdenes, anularlas e historial de órdenes, a cargo de los skills de operación conversacional; esta es el plano de pensamiento — tener claro cómo fijar los pesos, cómo leer los resultados del backtest y qué significan los parámetros de riesgo; toda ejecución sigue pasando por el servicio de paper trading. Solo paper: sin dinero real; el backtest y el paper run no anticipan el futuro." }, "fr": { "name": "Wire · Laboratoire de stratégies", "summary": "Quatre skills de Wire · Laboratoire de stratégies : conception de stratégie (sémantique des poids et validation du universe), interprétation des backtests (cinq exigences et trois interdits), paramètres de maîtrise des risques (sens / magnitude / limites) et supervision du paper run (événements d'exécution / instantanés de poids / manques de données). Répartition avec la suite de paper trading de Wire : celle-là est le plan opérationnel — ouvrir des comptes, passer et annuler des ordres, historique des ordres, assurés par les skills d'opération conversationnelle ; celle-ci est le plan de la réflexion — savoir comment fixer les poids, lire les résultats d'un backtest, ce que signifient les paramètres de risque ; toute exécution passe toujours par le service de paper trading. Paper uniquement : pas d'argent réel ; backtest et paper run n'annoncent pas l'avenir." }, "ja": { "name": "Wire · ストラテジーラボ", "summary": "Wire ストラテジーラボの四技能セット：ストラテジー設計（ウェイトの意味論と universe の検証）、バックテスト解読（五つの要求事項と三つの禁止事項）、リスク管理パラメータ（意味／規模感／限界）、ペーパーラン監視（実行イベント／ウェイトのスナップショット／データ欠損）。Wire ペーパートレードスイートとの役割分担：あちらは操作面——口座開設・発注・取消・注文履歴は対話操作スキルが担います。こちらは思考面——ウェイトをどう定めるか、バックテスト結果をどう読むか、リスク管理パラメータが何を意味するかを明確にします。約定はすべて引き続きペーパートレードサービス経由です。paper-only：実資金なし。バックテストとペーパーランは未来を予告しません。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "strategy-design", "description": "柏讯策略工坊 · 策略设计。当用户想把一个投资想法变成可回测的多标的策略（universe + 目标权重 + 调仓频率）、想理解权重语义（和为 1、非负、余量现金）、想取舍调仓频率与换手、或想知道哪些表达不被支持（做空/杠杆/盘中调仓）时使用。产出：一份可创建的 weight 模式策略 spec。" }, { "name": "strategy-backtest-interpretation", "description": "柏讯策略工坊 · 回测解读。当用户拿到一份回测结果（尤其 walk-forward 分窗、开启成本模型、带基准对比、参数扫描）想知道\"这个数字说明什么\"、要求解读权益曲线与统计、或问\"这能当作未来收益预期吗\"时使用。核心纪律：历史模拟不预示未来；单次回测不下有效结论；不忽略成本与基准。" }, { "name": "strategy-risk-controls", "description": "柏讯策略工坊 · 风控参数。当用户问\"单票最多买多少\"\"止损设多少\"\"换手太高怎么办\"\"设了会自动执行吗\"——即想理解单票上限/集中度/换手/止损/冷却期这些风控参数的含义、量级参考、代价与局限时使用。核心纪律：风控参数是策略层参数，不是引擎保证。" }, { "name": "strategy-paper-supervision", "description": "柏讯策略工坊 · 模拟盘督导。当用户已启动权重模式的纸面实盘（paper run），想知道\"为什么没下单\"\"权重怎么和目标对不上\"\"缩量是什么意思\"\"停了会平仓吗\"、需要查看运行事件与权重快照、或要区分信号未触发与数据缺口时使用。" } ], "mcpServers": [ { "registryName": "fd-paper-trading" }, { "registryName": "fd-quant" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "aMdkHzCbsc3fPsYiqMDUOA", "slug": "pack-amdkhzcbsc3fpsyiqmduoa", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "舆情", "公告", "wire" ], "canonical": { "brief": "Wire Sentiment, a three-skill set that turns announcement text into observable dimensions that carry verbatim quotes and can be verified and tracked — announcement sentiment review (sentiment-document-review: per-document scores must carry citations with character offsets, multiple documents yield a" }, "human": { "zh-CN": { "name": "柏讯 · 舆情驱动", "summary": "柏讯舆情驱动三件套：把公告文本变成带原文引用、可核验、可追踪的观察维度——公告舆情审读（sentiment-document-review，逐篇打分必带引用与字符定位、多篇给分布而非平均情绪、服务拒答如实转述、低置信显式标注）、舆情转向追踪（sentiment-trend-watch，按公告发布日排布、转向点必须附对应文档、样本门槛单期 ≥3 篇且高置信 ≥2 篇、覆盖面与置信度固定两栏）、舆情与价格持仓的描述性对照（sentiment-signal-bridge，只陈述时间先后不断言因果、相关性陈述须 ≥12 对且走计算面、严禁给买卖建议与「信号」称谓）。与投研套件的分工：研报的催化剂章引用本套件分数，不重复打分（同一服务同一幂等键，重复调用命中缓存）；与策略工坊的分工：那边做信号与回测，本套件只做描述性对照不给信号。分数不预示价格、当前版本（gold-v10 / prompt v7）两轮复测置信桶单调递增、方向一致率 99.19%，但五档精档之间仍有相邻档误判，均如实标注。" }, "en": { "name": "Wire · Sentiment-Driven", "summary": "Wire Sentiment, a three-skill set that turns announcement text into observable dimensions that carry verbatim quotes and can be verified and tracked — announcement sentiment review (sentiment-document-review: per-document scores must carry citations with character offsets, multiple documents yield a distribution rather than an average mood, service refusals are relayed as-is, low confidence is flagged explicitly), sentiment turn tracking (sentiment-trend-watch: scores laid out by announcement publication date, every turn point must cite its corresponding document, sample thresholds of ≥3 documents per period and ≥2 high-confidence, coverage and confidence as two fixed columns), and a descriptive comparison of sentiment against prices and holdings (sentiment-signal-bridge: states temporal order only and never asserts causality, correlation claims require ≥12 pairs and go through the compute plane, buy/sell advice and the word \"signal\" are strictly forbidden). Division of labor with the Equity Research suite: the report's catalysts section cites this suite's scores instead of re-scoring (same service, same idempotency key — repeat calls hit the cache); division of labor with the Strategy Lab: that side builds signals and backtests, this suite only does descriptive comparison and gives no signals. Scores do not predict prices; in the current version (gold-v10 / prompt v7) two rounds of re-testing show monotonically increasing confidence buckets and 99.19% directional agreement, yet adjacent-band confusion persists among the five fine bands — all stated as-is." }, "es": { "name": "Wire · Dinámica de la opinión pública", "summary": "Tríada de skills de Wire · Opinión pública: convierte el texto de los anuncios en dimensiones observables con cita literal, verificables y rastreables — lectura de sentimiento de anuncios (sentiment-document-review: cada puntuación lleva cita y posición de caracteres, varias piezas dan una distribución y no una media del ánimo, los rechazos del servicio se transmiten tal cual, la baja confianza se marca de forma explícita), seguimiento de los virajes de opinión (sentiment-trend-watch: puntuaciones ordenadas por fecha de publicación del anuncio, cada punto de viraje cita su documento correspondiente, umbral muestral de ≥3 piezas por periodo y ≥2 de alta confianza, cobertura y confianza como dos columnas fijas) y contraste descriptivo entre sentimiento, precios y posiciones (sentiment-signal-bridge: solo afirma el orden temporal y nunca la causalidad, hablar de correlación exige ≥12 pares y pasar por el plano de cálculo, queda terminantemente prohibido dar consejos de compra/venta o llamar «señal» a esto). Reparto con la suite de investigación de inversiones: la sección de catalizadores del informe cita las puntuaciones de esta suite y no vuelve a puntuar (mismo servicio, misma clave de idempotencia — las llamadas repetidas dan en la caché); reparto con el laboratorio de estrategias: allí se hacen señales y backtests, aquí solo hay contraste descriptivo y no se dan señales. Las puntuaciones no anticipan precios; en la versión actual (gold-v10 / prompt v7) dos rondas de retest muestran cubos de confianza monótonamente crecientes y un 99,19% de coincidencia direccional, pero entre las cinco bandas finas persisten confusiones entre bandas adyacentes — todo se declara tal cual." }, "fr": { "name": "Wire · Dynamique de l'opinion publique", "summary": "Trio de skills de Wire · Opinion publique : il transforme le texte des annonces en dimensions observables avec citation littérale, vérifiables et traçables — lecture du sentiment des annonces (sentiment-document-review : chaque score porte une citation avec position des caractères, plusieurs pièces donnent une distribution plutôt qu'une moyenne d'humeur, les refus du service sont relayés tels quels, la faible confiance est signalée explicitement), suivi des tournants d'opinion (sentiment-trend-watch : scores rangés par date de publication de l'annonce, chaque tournant cite le document correspondant, seuils d'échantillon de ≥3 pièces par période et ≥2 à haute confiance, couverture et confiance en deux colonnes fixes) et confrontation descriptive du sentiment aux prix et positions (sentiment-signal-bridge : seul l'ordre temporel est énoncé, jamais la causalité, parler de corrélation exige ≥12 paires et passe par le plan de calcul, toute recommandation d'achat/vente et l'appellation « signal » sont formellement interdites). Répartition avec la suite de recherche en investissement : la section catalyseurs du rapport cite les scores de cette suite sans renoter (même service, même clé d'idempotence — les appels répétés tombent dans le cache) ; répartition avec le laboratoire de stratégies : là-bas on fait signaux et backtests, ici seulement une confrontation descriptive, sans signaux. Les scores n'anticipent pas les prix ; dans la version actuelle (gold-v10 / prompt v7), deux rondes de retest montrent des buckets de confiance monotone croissante et 99,19 % de concordance directionnelle, mais des confusions entre bandes adjacentes subsistent parmi les cinq bandes fines — le tout est déclaré tel quel." }, "ja": { "name": "Wire · 世論ドリブン", "summary": "Wire 世論ドリブンの三技能セット：公告テキストを、原文引用付きで検証可能・追跡可能な観測次元に変えます——公告センチメント審読（sentiment-document-review：1 件ごとの採点に引用と文字位置が必須、複数件は平均センチメントでなく分布を提示、サービスの拒否はありのまま伝達、低信頼は明示）、センチメント転換の追跡（sentiment-trend-watch：公告の公開日に沿って採点を配置、転換点には対応する文書を必ず付記、サンプル要件は 1 期 ≥3 件かつ高信頼 ≥2 件、カバレッジと信頼度を固定の 2 列で報告）、センチメントと価格・保有の記述的対照（sentiment-signal-bridge：時間的な前後だけを述べて因果は断言しない、相関の記述には ≥12 ペアが必要で計算面を経由、売買の推奨と「シグナル」という呼称は厳禁）。投資リサーチスイートとの分担：レポートのカタリスト章は本スイートのスコアを引用し、再採点しない（同一サービス・同一冪等キーで、重複呼び出しはキャッシュにヒット）。ストラテジーラボとの分担：あちらはシグナルとバックテストを担当、本スイートは記述的対照のみでシグナルを出しません。スコアは価格を予告しません。現行版（gold-v10 / prompt v7）は 2 回の再試験で信頼度バケットが単調増加・方向一致率 99.19% ですが、5 つの精細バンドの間には隣接バンド同士の誤判定が残り、いずれもありのまま明記します。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "sentiment-document-review", "description": "柏讯舆情驱动 · 公告舆情审读。当用户要求\"看看这家公司最近的公告口径怎么样\"\"给这几篇公告做个舆情判断\"\"这些公告偏正面还是负面\"\"公告里有什么值得注意的表述\"时使用。取公告全文（巨潮面）→ 逐篇打分（fd-sentiment，分数必带原文引用）→ 呈现分数/band/置信度/引用片段与定位。多篇给分布而非平均情绪；拒答时如实转述；低置信显式标注。不做交易信号，不给情绪指数。" }, { "name": "sentiment-trend-watch", "description": "柏讯舆情驱动 · 舆情转向追踪。当用户要求\"这家公司最近舆情是在变好还是变差\"\"这段时间公告口径有没有转向\"\"按月看公告情绪的分布变化\"\"找出情绪转折点对应的公告\"时使用。按公告发布日排布分数，给逐期分布与转向点（附对应文档），并强制报告覆盖面与置信度两栏。样本不足时明示不下转向结论。不给情绪指数、不给交易信号。" }, { "name": "sentiment-signal-bridge", "description": "柏讯舆情驱动 · 舆情与价格/持仓的描述性对照。当用户要求\"舆情和股价是什么关系\"\"情绪是不是领先价格\"\"看看公告情绪和我持仓的对应\"\"舆情能不能当信号用\"时使用。做的是描述性对照：把分数分布的时间位置与价格/持仓的时间位置并排放，说清先后与是否同期，并强制标注样本是否足以做描述性陈述。严禁给买卖建议、严禁把舆情称为信号。价格通道或账本不可用时明示降级。" } ], "mcpServers": [ { "registryName": "fd-sentiment" }, { "registryName": "fd-cn-report" }, { "registryName": "fd-paper-trading" }, { "registryName": "fd-daas-mcp" }, { "registryName": "fd-quant" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "5mqV0jhihj0AhdNPCpc1iw", "slug": "pack-5mqv0jhihj0ahdnpcpc1iw", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "经济", "金融", "wire" ], "canonical": { "brief": "Regional economy comparison: a multi-indicator system backed by the city panel (209 variables × 297 cities × 2000-2024) and the statistical yearbook (31 provinces), supporting cross-sectional rankings and percentiles, growth-contribution structure decomposition, similar-city clustering and regional-" }, "human": { "zh-CN": { "name": "柏讯 · 区域经济套件", "summary": "区域经济比较：城市面板（209 变量 × 297 城 × 2000-2024）与统计年鉴（31 省）支撑的多维指标体系、横向排名分位、增长贡献结构分解、相似城市聚类与区域差距度量；配套统计检验方法。" }, "en": { "name": "Wire · Regional Economy Suite", "summary": "Regional economy comparison: a multi-indicator system backed by the city panel (209 variables × 297 cities × 2000-2024) and the statistical yearbook (31 provinces), supporting cross-sectional rankings and percentiles, growth-contribution structure decomposition, similar-city clustering and regional-gap measurement; with statistical testing methods included." }, "es": { "name": "Wire · Suite de economía regional", "summary": "Comparación de economía regional: sistema multi-indicador respaldado por el panel de ciudades (209 variables × 297 ciudades × 2000-2024) y el anuario estadístico (31 provincias), con rankings y percentiles transversales, descomposición estructural de la contribución al crecimiento, agrupamiento de ciudades similares y medición de brechas regionales; incluye métodos de contraste estadístico." }, "fr": { "name": "Wire · Suite économie régionale", "summary": "Comparaison économique régionale : système multi-indicateurs adossé au panel de villes (209 variables × 297 villes × 2000-2024) et à l'annuaire statistique (31 provinces), avec classements et centiles transversaux, décomposition structurelle des contributions à la croissance, regroupement de villes similaires et mesure des écarts régionaux ; méthodes de tests statistiques incluses." }, "ja": { "name": "Wire · 地域経済スイート", "summary": "地域経済比較：都市パネル（209 変数 × 297 都市 × 2000-2024）と統計年鑑（31 省）に支えられた多次元指標体系、横断ランキングと分位点、成長寄与の構造分解、類似都市のクラスタリング、地域格差の計測。統計的検定の手法を同梱。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "regional-economy-compare", "description": "区域经济比较：对中国城市/省份做多维指标横向排名、分位定位、增长结构分解与相似城市聚类。当用户要比较两个或多个城市/省份的经济实力、产业结构、财政与居民收入，要某城市在全省或全国的排名与分位，要拆解经济总量增长的产业贡献与拉动点数，或要按结构找与目标城市可比的对照组时使用。数据走柏讯城市面板与统计年鉴。" }, { "name": "statistical-inference", "description": "统计检验与探索性分析。当用户要求做描述统计与分布诊断、选假设检验（t/ANOVA/卡方/非参）、判断两组或城市间差异是否显著、做相关性与配对分析、诊断时序平稳性与单位根、做多重比较校正，或对已有的本地 CSV/表格做探索性分析时使用。走柏讯服务端序列取数，价格类序列走本地 DAAS。" } ], "mcpServers": [ { "registryName": "fd-cn-report" }, { "registryName": "fd-daas-mcp" }, { "registryName": "fd-find-data-business-mcp" }, { "registryName": "fd-open-data-mcp" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "mFYxWUfp5oHsPcpRDhxAhw", "slug": "pack-mfyxwufp5ohspcprdhxahw", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "经济", "wire" ], "canonical": { "brief": "Wire Portfolio Risk, a four-skill set: risk budgeting (per-position and sector caps + proportional reallocation on breach + an explicit iteration cap + honest reporting when it fails to converge, producing a compliant TargetWeights instance), drawdown and volatility checkup (max drawdown start/end, " }, "human": { "zh-CN": { "name": "柏讯 · 组合风险", "summary": "柏讯组合风险四件套：风险预算（单票与板块上限 + 超限按比例重分配 + 迭代上限显式 + 未收敛如实报告，产出合规 TargetWeights）、回撤与波动体检（最大回撤起止/年化波动/HHI/前 N 占比/行业集中度/有效标的数 + 基准超额与跟踪误差，指标走计算面，体检不预测）、止损纪律（量级起点 -5%/-10%/20 日冷却 + 按标的波动重标 + 三处明示引擎不自动执行 + 触发检查只呈现状态不新增判断）、再平衡方案（逐标的差额/方向/股数 + 佣金印花税滑点分项成本 + 可送权重模式回测 + 不含择时）。与决策复盘套件的分工：复盘看现状（集中度/回撤/偏离），本套件定纪律（预算怎么分、止损怎么设、什么时候该动手），一静一动。与策略工坊的分工：工坊讲通用参数含义，本套件把纪律应用到用户的实际持仓上。paper-only：无真实资金；所有指标是历史与当前状态描述，不构成投资建议。" }, "en": { "name": "Wire · Portfolio Risk", "summary": "Wire Portfolio Risk, a four-skill set: risk budgeting (per-position and sector caps + proportional reallocation on breach + an explicit iteration cap + honest reporting when it fails to converge, producing a compliant TargetWeights instance), drawdown and volatility checkup (max drawdown start/end, annualized volatility, HHI, top-N weight, sector concentration, effective number of holdings, plus benchmark excess return and tracking error — indicators come from the compute plane and a checkup predicts nothing), stop-loss discipline (magnitude anchors -5%/-10% and a 20-day cooldown + volatility-based re-scaling per holding + three explicit statements that the engine never auto-executes + trigger checks that present status without adding judgments), and a rebalancing plan (per-holding delta / direction / share counts + itemized commission, stamp tax and slippage costs + a weight rule submittable to weight-mode backtest + no market timing). Division of labor with the Decision Review suite: review looks at the present state (concentration / drawdown / drift); this suite sets the discipline (how to split the budget, where stops go, when to act) — one static, one active. Division of labor with the Strategy Lab: the lab explains generic parameter meanings; this suite applies the discipline to the user's actual holdings. Paper-only: no real money; every indicator describes history and current state and does not constitute investment advice." }, "es": { "name": "Wire · Riesgo de cartera", "summary": "Cuatro skills de Wire · Riesgo de cartera: presupuesto de riesgo (límites por posición y por sector + reasignación proporcional al excederlos + tope de iteraciones explícito + informe honesto si no converge, y produce una instancia de TargetWeights conforme desde su generación), chequeo de caída y volatilidad (inicio/fin de la caída máxima, volatilidad anualizada, HHI, peso del top N, concentración sectorial, número efectivo de posiciones, más exceso y error de seguimiento frente al índice — los indicadores van por el plano de cálculo y el chequeo no predice), disciplina de stop-loss (anclas de magnitud -5%/-10% y enfriamiento de 20 días + reescalado por volatilidad de cada posición + tres recordatorios de que el motor no ejecuta automáticamente + comprobaciones que presentan estado sin añadir juicio), y plan de rebalanceo (delta por posición / dirección / número de acciones + costes desglosados de comisión, impuesto de sello y slippage + regla de pesos para el backtest en modo pesos + sin temporización del mercado). Reparto con la suite de revisión de decisiones: la revisión mira el estado actual (concentración / caída / desviación), esta suite fija la disciplina (cómo repartir el presupuesto, dónde poner los stops, cuándo actuar) — una estática, otra activa. Reparto con el laboratorio de estrategias: aquel explica el significado genérico de los parámetros, esta suite aplica la disciplina a las posiciones reales del usuario. Solo paper: sin dinero real; todos los indicadores describen el historial y el estado actual y no constituyen asesoramiento de inversión." }, "fr": { "name": "Wire · Risque de portefeuille", "summary": "Quatre skills de Wire · Risque de portefeuille : budget de risque (plafonds par ligne et par secteur + réallocation proportionnelle en cas de dépassement + plafond d'itérations explicite + compte rendu honnête si non-convergence, produit une instance TargetWeights conforme dès la génération), check-up repli et volatilité (début/fin du repli maximal, volatilité annualisée, HHI, poids du top N, concentration sectorielle, nombre effectif de lignes, plus surperformance et erreur de suivi contre l'indice de référence — les indicateurs passent par le plan de calcul et le check-up ne prédit rien), discipline de stop-loss (ancres de magnitude -5%/-10% et période de refroidissement de 20 jours + re-calibrage selon la volatilité de chaque ligne + trois rappels que le moteur n'exécute pas automatiquement + contrôles de déclenchement qui présentent l'état sans ajouter de jugement), et plan de rééquilibrage (delta par ligne / sens / nombre d'actions + coûts détaillés commission, droit de timbre, slippage + règle de poids transmettable au backtest en mode poids + aucun market timing). Répartition avec la suite de revue des décisions : la revue observe l'état présent (concentration / repli / dérive), cette suite fixe la discipline (comment répartir le budget, où placer les stops, quand agir) — l'une statique, l'autre active. Répartition avec le laboratoire de stratégies : le laboratoire explique le sens générique des paramètres, cette suite applique la discipline aux positions réelles de l'utilisateur. Paper uniquement : pas d'argent réel ; tous les indicateurs décrivent l'historique et l'état présent et ne constituent pas un conseil en investissement." }, "ja": { "name": "Wire · ポートフォリオリスク", "summary": "Wire ポートフォリオリスクの四技能セット：リスク予算（銘柄・セクター上限＋超過時の比例再配分＋反復上限を明示＋未収束は正直に報告。生成時点でコンプライアンス適合の TargetWeights を出力）、ドローダウンとボラティリティの検診（最大ドローダウンの起止／年率ボラティリティ／HHI／上位 N 占比／セクター集中度／実効銘柄数＋ベンチマークとの超過収益とトラッキングエラー。指標は計算面経由、検診は予測しない）、ストップロス規律（規模感の起点 -5%／-10%／20 日のクールダウン＋銘柄ごとのボラティリティによる再設定＋エンジンが自動執行しないことを三箇所で明示＋発火チェックは状態の提示のみで判断を増やさない）、リバランス案（銘柄ごとの差額／方向／株数＋手数料・印紙税・スリッページの内訳コスト＋ウェイトモードのバックテストに投入可能＋タイミング判断は含まない）。意思決定レビュースイートとの分担：レビューは現状を見る（集中度／ドローダウン／乖離）、本スイートは規律を定める（予算の配り方、ストップの置き方、いつ動くか）——静と動。ストラテジーラボとの分担：ラボは汎用的なパラメータの意味を説き、本スイートは規律をユーザーの実際の保有に適用します。paper-only：実資金なし。全指標は履歴と現状の記述であり、投資助言ではありません。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "portfolio-budget", "description": "柏讯组合风险 · 风险预算。当用户要求\"资金怎么分到这些标的/行业\"\"单票最多放多少\"\"按上限重新分配一遍\"时使用。产出：一份**生成即合规**的 TargetWeights 契约实例（单票上限 + 板块上限 + 迭代重分配，迭代上限显式，未收敛如实报告）+ 约束满足表。只管权重结构，不管\"该投多少钱\"；现金靠 Σweights < 1 留出，引擎不会替你扣。" }, { "name": "portfolio-drawdown-review", "description": "柏讯组合风险 · 回撤与波动体检。当用户要求\"看看这个组合的回撤有多深\"\"波动多大\"\"是不是太集中了\"\"和沪深300比怎么样\"时使用。产出：组合级指标（最大回撤含峰谷日、年化波动、HHI、前 N 占比、板块集中度、有效标的数）+ 基准对比（超额与跟踪误差）。指标一律来自 fd-quant 计算面，不手算、不预测，缺失如实标注。" }, { "name": "portfolio-stop-discipline", "description": "柏讯组合风险 · 止损纪律。当用户要求\"止损设多少\"\"移动止损怎么定\"\"止损设了会自动执行吗\"\"帮我看看哪些已经跌破线了\"时使用。产出：一份用户既定的止损规则（绝对止损 / 移动止损 / 冷却期，含量级起点与按波动重标的方法与依据）+ 触发检查的状态呈现（当前值 vs 阈值 + 冷却剩余）。**本技能必须首先声明：这些是用户的纪律设定，不是引擎保证——引擎不会自动执行止损，权重模式只消费 weights_series，契约的 constraints 只有单票与板块两个上限字段，没有任何止损字段。**" }, { "name": "portfolio-rebalance-plan", "description": "柏讯组合风险 · 再平衡方案。当用户要求\"从现在的持仓调到这个目标权重，具体买什么卖什么\"\"每只调多少股\"\"调仓成本大概多少\"\"这套再平衡规则历史上表现如何\"时使用。产出：逐标的差额 / 方向 / 整数股数（向下取整，与权重模式引擎同口径）+ 成本预估（佣金 / 印花税 / 滑点分项，参数显式）+ 可送入权重模式回测的权重规则。**方案不含择时建议**；并明确引擎不自动执行止损、只消费 weights_series，cash_floor 只是权重上限校验而非独立持仓约束。" } ], "mcpServers": [ { "registryName": "fd-paper-trading" }, { "registryName": "fd-quant" }, { "registryName": "fd-daas-mcp" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "cjT2AW1QDGPGbvNIFmLgDg", "slug": "pack-cjt2aw1qdgpgbvnifmlgdg", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "经济", "金融", "wire" ], "canonical": { "brief": "Macro cycle and conditions analysis: a six-domain indicator framework (growth / inflation / money & credit / fiscal / external trade / property & consumption) + cycle positioning (output-inflation quadrants) + leading-coincident-lagging stratification + transmission-chain reasoning + turning-point i" }, "human": { "zh-CN": { "name": "柏讯 · 宏观研究套件", "summary": "宏观周期与景气分析：六域指标框架（增长/通胀/货币信用/财政/外贸/地产消费）+ 周期定位（产出-通胀四象限）+ 领先-同步-滞后分层 + 传导链推演 + 拐点识别；配套统计检验方法做相关性与平稳性验证。数据经柏讯开放数据面与统计年鉴实时检索，每个数值标注来源与数据期数。" }, "en": { "name": "Wire · Macro Research Suite", "summary": "Macro cycle and conditions analysis: a six-domain indicator framework (growth / inflation / money & credit / fiscal / external trade / property & consumption) + cycle positioning (output-inflation quadrants) + leading-coincident-lagging stratification + transmission-chain reasoning + turning-point identification, with statistical testing methods for correlation and stationarity checks. Data are retrieved in real time via the Wire open-data surface and statistical yearbooks, with every figure annotated by source and data period." }, "es": { "name": "Wire · Suite de Investigación Macro", "summary": "Análisis de ciclo macro y coyuntura: marco de indicadores de seis dominios (crecimiento / inflación / dinero y crédito / política fiscal / comercio exterior / inmobiliario y consumo) + posicionamiento cíclico (cuadrantes producción-inflación) + estratificación anticipados-coincidentes-retrasados + razonamiento de cadenas de transmisión + identificación de puntos de inflexión, con métodos de contraste estadístico para verificar correlación y estacionariedad. Los datos se recuperan en tiempo real por la superficie de datos abiertos de Wire y los anuarios estadísticos, y cada cifra se anota con fuente y período de datos." }, "fr": { "name": "Wire · Suite de Recherche Macro", "summary": "Analyse de cycle macro et de conjoncture : cadre d'indicateurs en six domaines (croissance / inflation / monnaie-crédit / finances publiques / commerce extérieur / immobilier-consommation) + positionnement cyclique (quadrants production-inflation) + stratification avancés-coïncidents-retardés + raisonnement en chaînes de transmission + identification des points d'inflexion, avec des méthodes de tests statistiques pour valider corrélation et stationnarité. Les données sont interrogées en temps réel via la surface de données ouvertes de Wire et les annuaires statistiques, chaque chiffre étant annoté de sa source et de sa période de données." }, "ja": { "name": "Wire · マクロ研究スイート", "summary": "マクロ周期と景気の分析：6領域の指標フレーム（成長/インフレ/金融・信用/財政/対外貿易/不動産・消費）+ 周期ポジショニング（産出-インフレの四象限）+ 先行・一致・後行の層別 + 伝播チェーンの推論 + 転換点の識別。相関と定常性の検証には統計検定の手法を併用。データはWireのオープンデータサーフェスと統計年鑑からリアルタイムに検索し、数値ごとに出典とデータ期数を注記。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "macro-cycle-analysis", "description": "宏观周期与景气分析。当用户要求判断当前经济周期位置（复苏/过热/滞胀/衰退）、做六域景气扫描（增长/通胀/货币信用/财政/外贸/地产消费）、做领先-同步-滞后指标分层、识别景气拐点、梳理宏观变量到其业务问题的传导链，或做省际补充与国际对比时使用。走柏讯服务端宏观数据，不含收益率曲线与互换分析。" }, { "name": "statistical-inference", "description": "统计检验与探索性分析。当用户要求做描述统计与分布诊断、选假设检验（t/ANOVA/卡方/非参）、判断两组或城市间差异是否显著、做相关性与配对分析、诊断时序平稳性与单位根、做多重比较校正，或对已有的本地 CSV/表格做探索性分析时使用。走柏讯服务端序列取数，价格类序列走本地 DAAS。" }, { "name": "china-macro-brief-workflow", "description": "中国宏观经济分析简报工作流（数据-中国经济包入口技能）。当用户询问中国宏观经济问题（GDP、通胀、货币、社融、地产、外贸、财政、行业景气），要求宏观数据查询、趋势分析、行业对照或宏观简报时使用。驱动从指标框架到图表简报的完整流程。" } ], "mcpServers": [ { "registryName": "fd-cn-report" }, { "registryName": "fd-daas-mcp" }, { "registryName": "fd-find-data-business-mcp" }, { "registryName": "fd-open-data-mcp" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "dxT2HEZ1z8GibrjMAgFJJw", "slug": "pack-dxt2hez1z8gibrjmagfjjw", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "经济", "wire" ], "canonical": { "brief": "Wire industry-chain linkage four-skill suite: chain mapping (multi-level upstream/midstream/downstream segments with representative names and attribution rationale, key variables and sources per segment, and two mapping-confidence tiers); transmission verification (per-segment correlations + lag sca" }, "human": { "zh-CN": { "name": "柏讯 · 产业链联动", "summary": "柏讯产业链联动四件套：链条映射（上中下游环节可多级、代表标的与归属依据、每环关键变量与来源、两档映射置信度）、传导验证（逐环相关性 + 时滞扫描 + 样本量与窗口，未验证环节明示不默认成立，不显著如实呈现，频度差处理，非因果表述）、环节受益分析（基于已验证的传导结论给候选，逐候选项回指传导依据，未验证环节不作依据，不预测价格）、链条景气跟踪（按环节分组的变量走势 + 来源与最新期数 + 缺口标注，无预测无建议）。与既有「行业链分析」技能的关系：那是分析技能，本套件是它的可安装套件化 + 可执行出口版本（从产业链关系走到可回测的权重），不是重复实现。核心纪律：产业链叙事最容易滑向讲漂亮故事——上游涨价到中游承压再下游提价听起来总是对的，但时滞、幅度与是否发生都必须有数据支撑；无数据的那一环明示未验证。传导结论是历史统计关系不是因果证明，不构成投资建议。" }, "en": { "name": "Wire · Industry-Chain Linkage", "summary": "Wire industry-chain linkage four-skill suite: chain mapping (multi-level upstream/midstream/downstream segments with representative names and attribution rationale, key variables and sources per segment, and two mapping-confidence tiers); transmission verification (per-segment correlations + lag scans + sample sizes and windows; unverified segments are stated as such and never assumed to hold, insignificant results reported honestly, frequency mismatches handled explicitly, no causal wording); segment-beneficiary analysis (candidates based on verified transmission conclusions, each candidate pointing back to its transmission evidence, unverified segments never used as a basis, no price forecasting); and chain-conditions tracking (variable trends grouped by segment + sources and latest periods + gap flags; no forecasts, no advice). Relation to the existing industry-chain-analysis skill: that is the analysis skill, this suite is its installable, executable-outlet form (from industry-chain relations to backtestable weights), not a duplicate implementation. Core discipline: industry-chain narratives slide most easily into pretty storytelling — upstream price rises squeezing midstream margins before downstream price hikes always sound right, but lags, magnitudes and whether it happened at all must be data-backed; segments without data are declared unverified. Transmission conclusions are historical statistical relations, not causal proof, and do not constitute investment advice." }, "es": { "name": "Wire · Enlace de la Cadena Industrial", "summary": "Suite de cuatro habilidades de enlace de la cadena industrial de Wire: mapeo de la cadena (segmentos aguas arriba/intermedios/aguas abajo de varios niveles, con valores representativos y fundamento de adscripción, variables clave y fuentes por segmento, y dos niveles de confianza de mapeo); verificación de la transmisión (correlaciones por segmento + barrido de desfases + tamaño de muestra y ventanas; los segmentos sin verificar se declaran y no se dan por ciertos, los resultados no significativos se presentan tal cual, las diferencias de frecuencia se tratan de forma explícita, sin lenguaje causal); análisis de beneficiarios por segmento (candidatos basados en conclusiones de transmisión verificadas, cada candidato remitiendo a su evidencia de transmisión, los segmentos sin verificar nunca como base, sin predicción de precios); y seguimiento de la coyuntura de la cadena (tendencias de variables agrupadas por segmento + fuentes y último período + faltantes señalados, sin pronósticos ni recomendaciones). Relación con la habilidad existente de análisis de cadena industrial: aquella es la habilidad de análisis y esta suite es su versión instalable con salida ejecutable (de las relaciones de la cadena a ponderaciones backtesteables), no una implementación duplicada. Disciplina central: las narrativas de cadena industrial derivan con facilidad en cuentos bonitos — que el precio suba arriba, apriete al medio y empuje abajo suena siempre convincente, pero desfases, magnitudes y si ocurrió deben estar respaldados por datos; el segmento sin datos se declara sin verificar. Las conclusiones de transmisión son relaciones estadísticas históricas, no prueba causal, y no constituyen asesoramiento de inversión." }, "fr": { "name": "Wire · Maillage de la Chaîne Industrielle", "summary": "Suite en quatre compétences de maillage de la chaîne industrielle de Wire : cartographie de la chaîne (segments amont/intermédiaire/aval multi-niveaux, valeurs représentatives et justification de rattachement, variables clés et sources par segment, deux niveaux de confiance de cartographie) ; vérification de la transmission (corrélation par segment + balayage des décalages + tailles d'échantillon et fenêtres ; les segments non vérifiés sont signalés et jamais tenus pour acquis, les résultats non significatifs restitués tels quels, les écarts de fréquence traités explicitement, sans formulation causale) ; analyse des bénéficiaires par segment (candidats fondés sur des conclusions de transmission vérifiées, chaque candidat renvoyant à sa preuve de transmission, segments non vérifiés jamais retenus comme base, pas de prévision de prix) ; et suivi de la conjoncture de la chaîne (trajectoires de variables groupées par segment + sources et dernières périodes + manquants signalés, sans prévision ni recommandation). Relation avec la compétence existante d'analyse de chaîne industrielle : celle-ci est la compétence d'analyse, cette suite en est la version installable à sortie exécutable (des relations de chaîne aux pondérations backtestables), pas une implémentation dupliquée. Discipline centrale : les récits de chaîne industrielle dérapent facilement vers le récit trop beau — l'amont qui monte, le milieu qui souffre, l'aval qui répercute sonne toujours juste, mais décalages, amplitudes et occurrence effective doivent être étayés par des données ; le segment sans données est déclaré non vérifié. Les conclusions de transmission sont des relations statistiques historiques, pas des preuves causales, et ne constituent pas un conseil en investissement." }, "ja": { "name": "Wire · 産業チェーン連動", "summary": "Wire産業チェーン連動の4技能スイート：チェーンマッピング（上流・中流・下流の多階層セグメント、代表銘柄と帰属根拠、セグメントごとの重要変数と出典、2段階のマッピング信頼度）、伝播検証（セグメントごとの相関 + ラグ走査 + サンプル数とウィンドウ。未検証セグメントは明示して既定成立としない、有意でない結果はありのまま提示、頻度差は明示的に処理、因果表現はしない）、セグメント受益分析（検証済みの伝播結論に基づく候補、候補ごとに伝播根拠へ回参照、未検証セグメントは根拠にしない、価格は予測しない）、チェーン景気トラッキング（セグメント別変数の推移 + 出典と最新期 + 欠損の注記。予測も助言もなし）。既存の産業チェーン分析スキルとの関係：あちらは分析スキルであり、本スイートはインストール可能で実行出口を持つ版（産業チェーン関係からバックテスト可能なウェイトまで）であり、重複実装ではない。中核規律：産業チェーンの語りは綺麗な物語に流れやすい——上流の値上がりが中流の圧迫を経て下流の値上げに至る話は常に正しく聞こえるが、ラグ・幅・そもそも起きたかどうかはデータで支える必要がある。データのないセグメントは未検証と明示する。伝播の結論は歴史的な統計関係であって因果証明ではなく、投資助言を構成しない。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "chain-mapping", "description": "柏讯产业链联动 · 链条映射。当用户要求\"把这个行业拆成上下游\"\"这几家公司分别在产业链哪一环\"\"这条链的结构和关键变量是什么\"时使用。产出：环节结构（上/中/下游，可多级）+ 代表标的（逐家给可解释的归属依据）+ 每环关键变量与来源 + 两档映射置信度（分类体系支持 / 公开资料推断）。只做结构映射，不做传导检验（交给 chain-transmission），不出候选（交给 chain-beneficiary）。" }, { "name": "chain-transmission", "description": "柏讯产业链联动 · 传导验证。当用户说\"上游涨价能不能传到中游毛利\"\"这条链的滞后有多长\"\"这个传导到底成不成立\"时使用。产出：逐环传导检验表——每环给相关系数、时滞扫描、样本量与窗口；无数据的环节明示\"未验证\"，不显著的结果如实呈现，频度差显式处理。结论是历史统计关系，不是因果证明。" }, { "name": "chain-beneficiary", "description": "柏讯产业链联动 · 环节受益分析。当用户说\"这条链里哪个环节该配\"\"毛利承压环节选哪几只\"\"把传导结论变成可回测的候选\"时使用。产出：可执行候选 universe 与目标权重（可进权重模式回测），每个候选回指某条已验证的传导结论；未验证环节不作依据。不预测价格、不给涨跌判断。" }, { "name": "chain-watch", "description": "柏讯产业链联动 · 链条景气跟踪。当用户要求\"盯一下这条链最近怎么样\"\"按环节给我看关键变量的最新走势\"\"哪些数据好久没更新了\"时使用。产出：按环节分组的变量走势表（来源 + 最新期数 + 缺口标注 + 变化方向），不含预测、不含建议。宿主 cron 可定时触发（本技能只给方法与输出形态，不实现调度）。" } ], "mcpServers": [ { "registryName": "fd-find-data-business-mcp" }, { "registryName": "fd-cn-report" }, { "registryName": "fd-quant" }, { "registryName": "fd-daas-mcp" }, { "registryName": "fd-paper-trading" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "XozWSsRjsA8MxFPkHBJZbw", "slug": "pack-xozwssrjsa8mxfpkhbjzbw", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "经济", "金融", "wire" ], "canonical": { "brief": "Industry cyclical conditions and chain transmission + corporate fundamentals and financial quality + panel econometrics: industry structure and upstream-downstream transmission grounded in listed-company evidence (GTA panel, 1,468 variables); DuPont decomposition and financial red-flag identificatio" }, "human": { "zh-CN": { "name": "柏讯 · 行业与企业套件", "summary": "行业景气与产业链传导 + 企业基本面与财务质量 + 面板计量方法：行业格局与上下游传导落到上市公司实证（GTA 面板 1,468 变量），企业端做杜邦分解与财务红旗项识别，计量侧支持固定效应、聚类稳健标准误与双重差分。" }, "en": { "name": "Wire · Industry & Corporate Suite", "summary": "Industry cyclical conditions and chain transmission + corporate fundamentals and financial quality + panel econometrics: industry structure and upstream-downstream transmission grounded in listed-company evidence (GTA panel, 1,468 variables); DuPont decomposition and financial red-flag identification on the corporate side; fixed effects, cluster-robust standard errors and difference-in-differences on the econometrics side." }, "es": { "name": "Wire · Suite de Industria y Empresa", "summary": "Coyuntura sectorial y transmisión por la cadena industrial + fundamentales corporativos y calidad financiera + econometría de datos de panel: la estructura sectorial y la transmisión aguas arriba/abajo se contrastan con evidencia de sociedades cotizadas (panel GTA, 1.468 variables); en lo corporativo, descomposición de DuPont y detección de señales de alerta financiera; en lo econométrico, efectos fijos, errores estándar robustos por clúster y diferencias en diferencias." }, "fr": { "name": "Wire · Suite Industrie et Entreprise", "summary": "Conjoncture sectorielle et transmission le long de la chaîne industrielle + fondamentaux d'entreprise et qualité financière + économétrie des données de panel : la structure sectorielle et la transmission amont/aval sont confrontées à des preuves tirées des sociétés cotées (panel GTA, 1 468 variables) ; côté entreprise, décomposition DuPont et détection des signaux d'alerte financière ; côté économétrie, effets fixes, erreurs-types robustes par cluster et différence de différences." }, "ja": { "name": "Wire · 産業・企業スイート", "summary": "業界の景気と産業チェーン伝播 + 企業のファンダメンタルズと財務品質 + パネル計量手法：業界の構図と上下流の伝播を上場企業の実証（GTAパネル、1,468変数）に落とし込み、企業側ではデュポン分解と財務レッドフラグの検出、計量側では固定効果・クラスタ頑健標準誤差・差分の差分（DID）に対応。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "industry-chain-analysis", "description": "行业景气与产业链传导分析。当用户要判断一个行业处在景气周期的哪个位置、梳理上下游投入产出与成本价格传导路径、用上市公司财务面板实证行业判断、刻画行业格局（规模/增速/集中度/进入壁垒），或把研报观点与公司真实财务数据对照时使用。走柏讯服务端行业与公司财务数据，不涉及价格与市值分析。" }, { "name": "corporate-fundamental-analysis", "description": "企业基本面与财务质量分析（柏讯线）。当用户要求给一家 A 股上市公司做三表速览、杜邦分解、现金流质量诊断、财务红旗项排查或同业分位对照时使用。走 GTA firm-year 财务面板 + 中国财报三表与附注，输出带来源与数据期数的可复核结论。本技能只做基本面质量判断，不做估值、不给目标价。" }, { "name": "panel-econometrics", "description": "面板数据计量分析。当用户要在\"个体 × 时间\"二维结构上做推断时使用——固定效应/随机效应与 Hausman 检验、聚类稳健标准误、动态面板（滞后因变量）、双重差分 DID 与平行趋势/事件研究检验、断点回归概念。面板取自柏讯服务端三类源：GTA 上市公司（firm-year）、中国城市（city-year）、世界银行 WDI（country-year），宏观控制变量取柏讯宏观序列。不涉及股价与市值建模。" } ], "mcpServers": [ { "registryName": "fd-cn-report" }, { "registryName": "fd-daas-mcp" }, { "registryName": "fd-find-data-business-mcp" }, { "registryName": "fd-open-data-mcp" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "ceuUUWYFYH_HiYnJ6UYqSg", "slug": "pack-ceuuuwyfyh-hiynj6uyqsg", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "经济", "金融", "wire" ], "canonical": { "brief": "Wire Policy Transmission four-skill set: policy identification (ministry items with source_type=ministry; read the body and extract the seven policy elements — issuing agency, document number, policy instrument, scope of application, effective date, enforcement horizon, quantified targets — each ele" }, "human": { "zh-CN": { "name": "柏讯 · 政策传导", "summary": "柏讯政策传导四件套：政策识别（部委条目 source_type=ministry，读正文抽发布机关/文号/政策工具/适用范围/生效时间/执行期限/量化目标七件套，每项要素附 item_id、as_of 与经 locate_evidence 核对的逐字片段）、行业映射（政策要素→受影响行业/区域→可检验变量三段映射，变量必须在年鉴/城市/公司面板真实检索到，两档置信度：政策明示/语义推断，无变量环节进未验证清单）、区域与企业传导（事件窗与政策强度序列两条路线逐环检验，相关系数、时滞扫描、样本量与窗口随行；结论三档——已验证/不显著/未验证环节，未验证环节必须显式标注，相关系数与样本量随每条结论输出；历史统计关系非因果证明）、政策跟踪（部委条目流计数与期间对比、健康度水位、已验证链路命中登记只转引不新增结论）。核心纪律：健康度前置（引用前先查 stream_health，lag 超阈值声明信息可能不完整）、区分已验证与未验证环节（无数据的一环明示未验证，绝不默认政策必然传导）、相关系数与样本量随结论输出、不因果表述不预测。不构成投资建议。" }, "en": { "name": "Wire · Policy Transmission", "summary": "Wire Policy Transmission four-skill set: policy identification (ministry items with source_type=ministry; read the body and extract the seven policy elements — issuing agency, document number, policy instrument, scope of application, effective date, enforcement horizon, quantified targets — each element with item_id, as_of and a verbatim quote verified through locate_evidence), industry mapping (three-stage mapping policy elements → affected industries/regions → testable variables; variables must be genuinely retrieved from the yearbook / city / firm panels, two confidence tiers — stated in the policy / inferred semantically — and any link without a variable enters the unverified list), regional and firm-level transmission testing (two routes — event windows and policy-intensity series — checked link by link, with correlation coefficients, lag scans, sample sizes and windows printed on each row; conclusions in three tiers — verified / not significant / unverified links, the latter explicitly labeled, correlation and sample size accompany every conclusion; historical statistical relations are not causal proof), and policy tracking (ministry stream counts and period comparison, health watermark, and a hit-register that merely cross-references verified chains without adding conclusions). Core discipline: health first (check stream_health before citing; over-threshold lag means declaring 'information may be incomplete'), separate verified from unverified links (a link with no data is explicitly unverified — never assume a policy must transmit), correlation and sample size accompany every conclusion, no causal wording, no forecasts. Not investment advice." }, "es": { "name": "Wire · Transmisión de políticas", "summary": "Suite Wire de transmisión de políticas en cuatro habilidades: identificación de políticas (ítems de ministerios con source_type=ministry; leer el cuerpo y extraer los siete elementos de política —organismo emisor, número de documento, instrumento, objeto y ámbito de aplicación, fecha de vigencia, plazo de ejecución, metas cuantificadas—, cada elemento con item_id, as_of y cita literal verificada con locate_evidence), mapeo sectorial (mapeo en tres etapas: elementos de política → industrias/regiones afectadas → variables contrastables; las variables deben buscarse de verdad en el anuario / panel urbano / panel de empresas, con dos niveles de confianza —establecida en la política / inferida semánticamente—, y todo eslabón sin variable entra en la lista de no verificados), contraste de transmisión regional y empresarial (dos rutas —ventanas de eventos y series de intensidad de política— verificadas eslabón a eslabón, con coeficientes de correlación, barrido de desfases, tamaño muestral y ventanas en cada fila; conclusiones en tres niveles —verificado / no significativo / eslabón no verificado, este último explícitamente rotulado, correlación y tamaño muestral en cada conclusión; las relaciones estadísticas históricas no son prueba causal), y seguimiento de políticas (conteos del flujo de ministerios y comparación entre periodos, marca de agua de salud, y un registro de aciertos que solo remite a cadenas verificadas sin añadir conclusiones). Disciplina central: salud primero (consultar stream_health antes de citar; retraso sobre el umbral → declarar «la información puede estar incompleta»), separar lo verificado de los eslabones no verificados (un eslabón sin datos queda explícitamente sin verificar; jamás asumir que la política transmite por fuerza), correlación y tamaño muestral con cada conclusión, sin lenguaje causal ni predicciones. No es asesoramiento de inversión." }, "fr": { "name": "Wire · Transmission des politiques", "summary": "Suite Wire « transmission des politiques » en quatre compétences : identification des politiques (items ministériels avec source_type=ministry ; lire le corps et extraire les sept éléments de politique — autorité émettrice, numéro de document, instrument, objet et périmètre d'application, date d'entrée en vigueur, horizon d'exécution, cibles quantifiées —, chaque élément avec item_id, as_of et citation verbatim vérifiée via locate_evidence), cartographie sectorielle (cartographie en trois temps : éléments de politique → industries/régions touchées → variables testables ; les variables doivent être réellement retrouvées dans l'annuaire / le panel urbain / le panel d'entreprises, deux niveaux de confiance — stipulé dans la politique / inféré sémantiquement —, et tout maillon sans variable rejoint la liste des non vérifiés), test de transmission régionale et d'entreprise (deux routes — fenêtres d'événements et séries d'intensité de politique — vérifiées maillon par maillon, avec coefficients de corrélation, balayage des retards, tailles d'échantillon et fenêtres sur chaque ligne ; conclusions en trois niveaux — vérifié / non significatif / maillon non vérifié, ce dernier explicitement signalé, corrélation et taille d'échantillon avec chaque conclusion ; les relations statistiques historiques ne prouvent pas la causalité), et suivi des politiques (comptages du flux ministériel et comparaison entre périodes, filigrane de santé, registre de correspondances qui renvoie aux chaînes vérifiées sans ajouter de conclusions). Discipline centrale : santé d'abord (consulter stream_health avant de citer ; retard au-delà du seuil → déclarer « l'information peut être incomplète »), distinguer le vérifié des maillons non vérifiés (un maillon sans données est explicitement non vérifié — ne jamais présumer qu'une politique transmet forcément), corrélation et taille d'échantillon avec chaque conclusion, aucun vocabulaire causal ni prévision. Ne constitue pas un conseil en investissement." }, "ja": { "name": "Wire · 政策伝達", "summary": "Wire 政策伝達 4 技能セット：政策識別（省庁項目 source_type=ministry を対象に本文を読み、政策要素 7 項目——公表機関・文書番号・政策手段・適用対象と範囲・発効時期・執行期限・数量目標——を抽出し、各要素に item_id、as_of、locate_evidence で照合した原文の逐語引用を付す）、業界マッピング（政策要素→影響を受ける業界・地域→検証可能な変数の 3 段マッピング。変数は年鑑／都市パネル／企業パネルで実際に検索できなければならず、信頼度は「政策に明記」「意味的推論」の 2 段階。変数のないリンクは未検証リストへ）、地域・企業伝達の検定（イベント窓と政策強度系列の 2 経路をリンクごとに検証：相関係数、ラグ走査、サンプル数、窓を各行に添え、結論は「検証済み／有意でない／未検証リンク」の 3 段階——未検証リンクは必ず明示し、相関係数とサンプル数を各結論に添える。歴史的な統計関係は因果証明ではない）、政策追跡（省庁項目フローの計数と期間対比、健全性ウォーターマーク、検証済みチェーンへの該当は転記のみで新たな結論を出さない）。中核規律：健全性の前置（引用前に stream_health を確認、閾値超の遅延では「情報が不完全な可能性」を明示）、検証済みと未検証リンクの区別（データのないリンクは明示的に未検証——政策が必ず伝わると想定しない）、相関係数とサンプル数を結論に添える、因果的表現・予測をしない。投資助言ではない。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "policy-detection", "description": "柏讯政策传导 · 政策识别。当用户要求\"最近有什么新政策\"\"把某部委近一月的发布理一下\"\"这条政策文件说了什么\"时使用。先查来源健康度，再取部委条目（source_type=ministry），读正文抽取政策要素（发布机关/文号/政策工具/适用对象与范围/生效时间/执行期限/量化目标），每项要素附 item_id、as_of 与经 locate_evidence 核对的原文逐字片段。只做识别与要素抽取，不做行业映射（交 policy-mapping）与传导检验（交 policy-transmission）。" }, { "name": "policy-mapping", "description": "柏讯政策传导 · 行业映射。当用户要求\"这条政策影响哪些行业\"\"政策落到哪些可检验的指标上\"\"给我政策—行业—变量的映射表\"时使用。把政策要素映射到受影响行业/区域维度，并逐一检索真实在册的检验变量（年鉴/城市面板/公司面板），两档置信度（政策明示 / 语义推断）逐条标注。只做映射：不做统计检验（交 policy-transmission），不出投资候选。" }, { "name": "policy-transmission", "description": "柏讯政策传导 · 区域与企业传导检验。当用户要求\"这条政策到底传没传到行业/区域数据\"\"政策发布后地方指标变了吗\"\"政策与公司科目有没有统计关系\"时使用。对映射交来的每条\"政策→变量\"假设逐环检验：相关系数、时滞扫描、样本量与窗口；结论三档——已验证 / 不显著 / 未验证环节，未验证环节必须显式标注，相关系数与样本量随每条结论输出。历史统计关系，不是因果证明，不预测。" }, { "name": "policy-watch", "description": "柏讯政策传导 · 政策跟踪。当用户要求\"帮我盯着部委发布\"\"每旬把新政策理一遍\"\"跟踪表里标一下哪些撞上了已验证链路\"时使用。按部委来源与政策工具类型筛条目流，输出条目清单、来源计数、期间对比与健康度水位；对映射/传导已登记的政策标\"同一事项的新披露\"，对撞上已验证链路要素的新政策只做\"命中登记\"不新增结论。不预测、不推荐；定时属宿主 cron。" } ], "mcpServers": [ { "registryName": "fd-info" }, { "registryName": "fd-find-data-business-mcp" }, { "registryName": "fd-quant" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "zxuH617BsFnj1QQk81QY-A", "slug": "pack-zxuh617bsfnj1qqk81qy-a", "category": "facade", "marketCategory": "legal", "tags": [ "识律", "法律", "lex" ], "canonical": { "brief": "A five-tool legal research set: statute and validity checks, similar-case retrieval reports, legal concept maps, five-corpus search with caliber comparison (departmental rules / treaties / discipline-inspection / case libraries / Party regulations), and validity lifecycle tracking. Data from the Lex" }, "human": { "zh-CN": { "name": "识律 · 法律检索套件", "summary": "法律检索研究五件套：法条与效力核查、类案检索报告、法律概念图解、五源语料检索与口径对照（规章/条约/纪检监察/案例库/党内法规）、法规效力生命周期追踪。数据来自识律 lex 线法规库（32,296 部）、裁判文书库（3,700 万篇）、指导案例库（278+15）与五源语料（47,261 篇），逐结果溯源。注：文书库不支持聚合统计，类案类产出为「检索式证据」口径而非统计结论。" }, "en": { "name": "Lex · Legal Research Suite", "summary": "A five-tool legal research set: statute and validity checks, similar-case retrieval reports, legal concept maps, five-corpus search with caliber comparison (departmental rules / treaties / discipline-inspection / case libraries / Party regulations), and validity lifecycle tracking. Data from the Lex law library (32,296), judgment library (37M), guiding cases (278+15) and five-corpus (47,261), with per-result provenance. Note: the judgment library supports no aggregate statistics — similar-case outputs are retrieval-based evidence, not statistical conclusions." }, "es": { "name": "Lex · Suite de investigación jurídica", "summary": "Cinco herramientas de investigación jurídica: verificación de artículos y vigencia, informes de casos análogos, mapas de conceptos jurídicos, búsqueda en cinco corpus con comparación de criterios (normas departamentales / tratados / disciplina-inspección / casos / normativas del Partido) y seguimiento del ciclo de vigencia. Datos de la biblioteca Lex (32.296 leyes), de sentencias (37 M), casos guía (278+15) y cinco corpus (47.261), con trazabilidad por resultado. Nota: la biblioteca de sentencias no admite estadísticas agregadas; los casos análogos son evidencia de búsqueda, no estadística." }, "fr": { "name": "Lex · Suite de recherche juridique", "summary": "Cinq outils de recherche juridique : vérification des textes et de la validité, rapports d'affaires analogues, cartes de concepts juridiques, recherche dans cinq corpus avec comparaison des critères (règlements ministériels / traités / discipline-inspection / cas / textes du Parti) et cycle de validité des textes. Données de la bibliothèque Lex (32 296 textes), de la base de décisions (37 M), des affaires guide (278+15) et des cinq corpus (47 261), traçabilité par résultat. Remarque : pas de statistiques agrégées sur la base de décisions ; les affaires analogues donnent des preuves de recherche, pas des statistiques." }, "ja": { "name": "Lex · リーガルリサーチスイート", "summary": "法令リサーチ五点セット：法条と効力の検証、類似事案検索レポート、法律概念の構造化図解、五ソース語料の検索と口径対照（部門規章／条約／紀検監察／案例庫／党内法規）、法規効力のライフサイクル追跡。データは Lex 法規ライブラリ（32,296 部）、裁判文書庫（3,700 万篇）、指導案例庫（278+15）、五ソース語料（47,261 篇）から、結果ごとに由来を追跡。注：裁判文書庫は集計統計に非対応で、類似事案系の出力は「検索型エビデンス」の口径であり統計的結論ではない。" } }, "whenToUse": "Browse under \"legal\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "statute-validity-check", "description": "法条检索与效力核查。当用户要查某法条原文、核对某条文是否现行有效、确认条号是否因修订变动、查某法律问题的法律依据（\"能不能\"\"是否有效\"\"如何认定\"\"依据哪一条\"）时使用。语义召回候选条文后逐条锚定效力状态，输出带效力状态与 provenance 的法条清单；非\"有效\"版本明确标注不得作现行法。" }, { "name": "case-retrieval-report", "description": "类案检索与裁判规则报告。当用户要检索类案、出案例检索报告、查某类纠纷的裁判规则或裁判倾向、做同案同判分析、找支持某主张的判例时使用。走裁判文书库 + 指导案例库多组关键词换搜，输出含真实案号、法院、裁判日期与 provenance 的检索报告。" }, { "name": "legal-concept-map", "description": "法律概念结构化图解。当用户要解释某个法律概念/制度（\"什么是善意取得\"\"表见代理怎么认定\"\"情势变更的适用条件\"）、要拆构成要件与例外、要找关联条文与权威解读时使用。以条级引用解析锚定条文，以五源语料补权威解读，输出要件—例外—关联条文的结构化图解。" }, { "name": "legal-corpus-explorer", "description": "五源法律语料检索与口径对照。当用户要查部门规章/地方规章、条约、纪检监察法规、人民法院案例库案例、党内法规，或要在多个语料源之间对照同一问题的规范口径、找某领域除法律外的细化规定（\"部门规章怎么规定的\"\"有没有条约\"\"党内法规怎么要求\"\"案例库里类似案件\"）时使用。按封闭词表指定 source 逐源检索，逐条带 provenance；不做跨源统计与\"权威解读占比\"类结论。" }, { "name": "statute-lifecycle-tracker", "description": "法规效力生命周期追踪与所在层级查证。当用户要了解某部法规的效力状态（有效/已修改/已废止/尚未生效/未知）、核对一部法规是否被后续修法改动、盘点某一领域的现行有效法规存量、或要判断\"某个规定还有效吗\"\"这部法改过几版\"时使用。以效力状态为权威口径逐条核验；如实声明发布日期与失效日期字段不可用，不做趋势推断。" } ], "mcpServers": [ { "registryName": "fd-legal-search-mcp" }, { "registryName": "fd-find-data-business-mcp" } ] } }, { "id": "DydB2cal-M9l91IMJYHshA", "slug": "pack-dydb2cal-m9l91imjyhsha", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "交易", "模拟盘", "量化", "wire" ], "canonical": { "brief": "Wire paper trading three-skill suite: conversational operations (accounts/positions/orders/cancellations), strategy workflow (templates and sandboxed custom code, backtest interpretation, paper-live runs) and market-data inspection (search/historical candlesticks/coverage boundaries stated honestly)" }, "human": { "zh-CN": { "name": "柏讯 · 模拟交易套件", "summary": "柏讯模拟交易（paper trading）三件套：对话操作（账户/持仓/下单/撤单）、策略工作流（模板与沙箱自定义代码、回测解读、纸面实盘）、行情检查（搜索/历史K线/覆盖边界诚实呈现）。paper-only：模拟盘、无真实资金；调用者身份逐请求鉴别，只操作自己的账户；行情覆盖由部署的概念族决定，技能以诚实呈现覆盖边界为第一纪律。" }, "en": { "name": "Wire · Paper Trading Suite", "summary": "Wire paper trading three-skill suite: conversational operations (accounts/positions/orders/cancellations), strategy workflow (templates and sandboxed custom code, backtest interpretation, paper-live runs) and market-data inspection (search/historical candlesticks/coverage boundaries stated honestly). Paper-only: simulated account, no real money; the caller's identity is authenticated per request and only their own accounts are touched; market-data coverage is set by the deployed concept families, and the skills' first discipline is to present coverage limits honestly." }, "es": { "name": "Wire · Suite de Trading Simulado", "summary": "Suite de tres habilidades de trading simulado de Wire: operaciones por conversación (cuentas/posiciones/órdenes/cancelaciones), flujo de trabajo de estrategias (plantillas y código propio en sandbox, interpretación del backtest, ejecución en vivo simulada) e inspección de cotizaciones (búsqueda/velas históricas/límites de cobertura presentados con honestidad). Solo simulación: cuenta simulada, sin dinero real; la identidad de quien llama se autentica en cada solicitud y solo se tocan sus propias cuentas; la cobertura de cotizaciones la deciden las familias de conceptos desplegadas, y la primera disciplina de las habilidades es exponer los límites de cobertura con honestidad." }, "fr": { "name": "Wire · Suite de Trading Fictif", "summary": "Suite de trois compétences de trading fictif de Wire : opérations conversationnelles (comptes/positions/ordres/annulations), flux de travail de stratégies (modèles et code personnalisé en bac à sable, interprétation du backtest, exécution en direct simulée) et inspection des cours (recherche/chandeliers historiques/limites de couverture présentées honnêtement). Simulation uniquement : compte simulé, sans argent réel ; l'identité de l'appelant est authentifiée à chaque requête et seuls ses propres comptes sont manipulés ; la couverture des cours est déterminée par les familles de concepts déployées, et la première discipline des compétences est de présenter honnêtement les limites de couverture." }, "ja": { "name": "Wire · ペーパートレードスイート", "summary": "Wireペーパートレードの3技能スイート：対話操作（口座/ポジション/発注/注文取消）、ストラテジーワークフロー（テンプレートとサンドボックスのカスタムコード、バックテスト結果の解釈、ペーパー実運転）、市況チェック（検索/履歴ローソク足/カバレッジ境界の誠実な提示）。ペーパートレード限定：模擬口座で実資金なし。呼び出し者の身元はリクエストごとに認証され、自分の口座のみ操作可能。市況カバレッジはデプロイされた概念ファミリーで決まり、技能の第一規律はカバレッジの境界を誠実に提示すること。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "paper-trading-conversation", "description": "柏讯模拟交易（paper trading）对话操作。当用户要求查看纸面账户与余额、开新模拟账户、查持仓明细（数量/成本/浮盈亏）、下单买入或卖出（市价/限价）、撤单、查订单历史，或询问某标的当前价格时使用。走柏讯模拟交易服务（paper-only 模拟盘，无真实资金、无真实订单），调用者身份逐请求鉴别——只操作自己的账户。" }, { "name": "paper-trading-strategy", "description": "柏讯模拟交易策略工作流。当用户要求创建量化策略（SMA 交叉/RSI 反转/布林突破/MACD 交叉/定投等模板，或自定义 on_bar 代码）、跑历史回测并解读权益曲线与统计（收益/回撤/夏普/胜率）、在模拟账户上启动或停止纸面实盘运行、查看运行历史时使用。paper-only：策略只在模拟盘上下单。" }, { "name": "paper-trading-market-inspection", "description": "柏讯模拟交易行情检查。当用户要求搜索某个可交易标的（按代码或名称）、查看某标的的历史 K 线（OHLCV）、了解当前部署行情面覆盖哪些标的类型、或为下单/回测确认某标的数据是否可用时使用。行情来自柏讯模拟交易服务的行情面（概念族可配，默认基金净值），覆盖面有限——本稿核心职责是诚实呈现覆盖范围。" } ], "mcpServers": [ { "registryName": "fd-paper-trading" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "I5n-sRpbXuFu6WLbh6PfvQ", "slug": "pack-i5n-srpbxufu6wlbh6pfvq", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "投委会", "辩论", "wire" ], "canonical": { "brief": "Wire IC Debate, a three-skill set: it turns the company-level debate procedure that TradingAgents/FinRobot spell out but never wire up into an installable pattern — a bull case (ic-bull-case: numbers may only cite evidence-set evidence_id, nothing self-invented or estimated, no arguing the opponent'" }, "human": { "zh-CN": { "name": "柏讯 · 投委会", "summary": "柏讯投委会三件套：把 TradingAgents/FinRobot 写好但未接线的公司层级辩论程序做成可安装的模式——多头立论（ic-bull-case，数字只能引用证据集 evidence_id、不自造不估算、不替对手立论、把己方理由说尽）、空头立论（ic-bear-case，对称规则 + 不扭曲多头已陈述证据）、裁决（ic-adjudication，入口技能：调度双方后按四纪律给出 Rating 契约实例——冲突不给 hold 须指强弱、证据不足给 review 不降级、裁决不引用任何数字、change_my_mind 必填），并落账至决策账本（结论可被 P5 结算——这是上游两家都缺失的闭环）。诚实定位：单 agent 分角色形态，三个角色由同一模型扮演，不构成独立视角；升级形态（萬星 A2A 三 persona 串行委派，深度 3 / 并发 1 / 分钟计费）已写入技能文本属产品演进。辩论不宣称提高预测准确率，是否更准由账本结算回答。裁决为研究参考，不构成投资建议。" }, "en": { "name": "Wire · IC Debate", "summary": "Wire IC Debate, a three-skill set: it turns the company-level debate procedure that TradingAgents/FinRobot spell out but never wire up into an installable pattern — a bull case (ic-bull-case: numbers may only cite evidence-set evidence_id, nothing self-invented or estimated, no arguing the opponent's side, your own case argued to its full extent), a bear case (ic-bear-case: symmetric rules, plus never distorting evidence the bull side has already stated), and an adjudication (ic-adjudication, the entry skill: after dispatching both sides it issues a Rating contract instance under four disciplines — a conflict never yields hold without stating which side is stronger and why, insufficient evidence yields review rather than a downgraded verdict, the verdict cites no numbers, change_my_mind is mandatory) — and then records the outcome to the decision ledger (conclusions can be settled by P5: the closed loop both upstream projects lack). Honest positioning: single-agent role-split form; the three roles are played by one model and are not independent perspectives; the upgrade path (Wanxing A2A serial delegation across three personas, depth 3 / concurrency 1 / per-minute billing) is written into the skill text as product evolution. The debate claims no gain in forecast accuracy — whether it is more accurate is answered by ledger settlement. Verdicts are research reference, not investment advice." }, "es": { "name": "Wire · Comité de inversión", "summary": "Tríada de skills de Wire · Comité de inversión: convierte en patrón instalable el procedimiento de debate a nivel de empresa que TradingAgents/FinRobot describen bien pero nunca cablean — argumento alcista (ic-bull-case: las cifras solo citan evidence_id del conjunto de evidencia, nada inventado ni estimado, sin argumentar por el rival y agotando los motivos propios), argumento bajista (ic-bear-case: reglas simétricas y sin distorsionar la evidencia ya expuesta por el lado alcista) y veredicto (ic-adjudication, skill de entrada: tras convocar a ambas partes emite una instancia del contrato Rating bajo cuatro disciplinas — el conflicto no da hold sin señalar qué lado es más fuerte y por qué, la evidencia insuficiente da review en vez de rebajar, el veredicto no cita cifras, change_my_mind obligatorio) — y asienta el resultado en el registro de decisiones (las conclusiones pueden liquidarse con P5: el circuito cerrado que falta en ambos proyectos de origen). Posicionamiento honesto: agente único con papeles repartidos; los tres roles los interpreta un mismo modelo y no son perspectivas independientes; la forma ampliada (delegación en serie de tres persona vía A2A de Wanxing, profundidad 3 / concurrencia 1 / facturación por minuto) figura ya en el texto de los skills como evolución del producto. El debate no alega mejorar la precisión de las previsiones: si lo hace lo responde la liquidación del registro. El veredicto es referencia de investigación, no asesoramiento de inversión." }, "fr": { "name": "Wire · Comité d'investissement", "summary": "Trio de skills de Wire · Comité d'investissement : il transforme en schéma installable la procédure de débat au niveau de l'entreprise que TradingAgents/FinRobot décrivent bien sans jamais la câbler — thèse haussière (ic-bull-case : les chiffres ne citent que les evidence_id du jeu de preuves, rien d'inventé ni d'estimé, pas d'argument pour le camp adverse, les raisons propres épuisées jusqu'au bout), thèse baissière (ic-bear-case : règles symétriques, sans déformer les preuves déjà énoncées par le camp haussier) et arbitrage (ic-adjudication, skill d'entrée : après avoir convoqué les deux parties, il produit une instance du contrat Rating sous quatre disciplines — un conflit ne donne jamais de hold sans désigner le camp le plus solide et pourquoi ; preuves insuffisantes → review plutôt qu'un abaissement ; l'arbitrage ne cite aucun chiffre ; change_my_mind obligatoire) — puis consigne le résultat au registre de décisions (les conclusions peuvent être réglées par P5 : la boucle fermée qui manque aux deux projets amont). Position honnête : forme mono-agent à rôles partagés ; les trois rôles sont joués par un même modèle et ne constituent pas des perspectives indépendantes ; la forme évoluée (délégation en série de trois persona via l'A2A de Wanxing, profondeur 3 / concurrence 1 / facturation à la minute) figure déjà dans le texte des skills comme évolution produit. Le débat ne prétend pas améliorer la précision des prévisions — c'est au règlement par le registre de le dire. L'arbitrage vaut comme référence de recherche, non comme conseil en investissement." }, "ja": { "name": "Wire · 投資委員会", "summary": "Wire 投資委員会の三技能セット：TradingAgents/FinRobot が丁寧に書きながら接線していない企業レベルの弁論手順を、インストール可能なパターンに仕立てます——強気の立論（ic-bull-case：数値は根拠セットの evidence_id の引用のみ、自作・概算なし、相手側の代弁なし、自陣の論拠は言い尽くす）、弱気の立論（ic-bear-case：対称の規則＋強気側が既に述べた根拠を歪めない）、裁定（ic-adjudication、入口スキル：双方を呼び出した後、四つの規律で Rating 契約インスタンスを出す——衝突で hold を出さず強弱と理由を指摘、根拠不足は review で格下げせず、裁定は数値を引用せず、change_my_mind は必須）——その上で結論を意思決定台帳に記帳（結論は P5 による決着が可能。上流の二プロジェクトのいずれにも欠けていた閉ループです）。正直な位置づけ：単一エージェントの役割分担形で、三つの役割を同じモデルが演じるため独立の視点ではありません。拡張形（Wanxing A2A による 3 persona の逐次委譲、深さ 3／同時実行 1／分単位課金）は製品の進化形として技能本文に記載済み。弁論は予測精度の向上を主張せず、より正確かは台帳の決着が答えます。裁定は研究参考であり、投資助言ではありません。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "ic-adjudication", "description": "柏讯投委会 · 裁决（入口技能）。当用户要求\"开投委会\"\"对某只股票做一次多空辩论并给结论\"\"裁决多头空头的分歧\"\"把辩论结论落账\"时使用。本技能是套件入口：调度 ic-bull-case 与 ic-bear-case 双方立论，然后按四条裁决纪律给出 Rating 契约实例（冲突不给 hold 须指强弱、证据不足给 review 不降级、裁决不引用任何数字、change_my_mind 必填），并完成落账（账本不可用时明示\"未落账\"）。单 agent 分角色形态，不构成独立视角。" }, { "name": "ic-bull-case", "description": "柏讯投委会 · 多头立论。当用户要求对某只股票构建看多论证、写多头观点、准备投委会辩论的多方意见、把证据集收敛成\"为什么该买\"的最强论证，或作为 ic-adjudication 的立论阶段被调度时使用。本技能不取数、不计算、不裁决——数字只能引用证据集的 evidence_id，论证组织是唯一职责。与 ic-bear-case 对称，空头未发言时不得替对手立论。" }, { "name": "ic-bear-case", "description": "柏讯投委会 · 空头立论（辩论三角色之一，与 ic-bull-case 完全对称）。当用户在投委会辩论流程中需要空头方论证、问「有什么理由不买」「多头论证哪里站不住」「估值凭什么下修」，或作为 ic-adjudication 的立论阶段被调度时使用。本技能不取数、不计算、不裁决——数字只能引用证据集的 evidence_id；不同意多头时只能提出新证据或指出其适用边界，不得改写数字。空头论证为空是合法产出，不得为对称而编造。" } ], "mcpServers": [ { "registryName": "fd-cn-report" }, { "registryName": "fd-daas-mcp" }, { "registryName": "fd-quant" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "kimcQr9qSyGmbXej3CmeDg", "slug": "pack-kimcqr9qsygmbxej3cmedg", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "经济", "wire" ], "canonical": { "brief": "Wire factor-selection four-skill suite: factor library and methodologies (a list of 52 factors in nine categories, with per-factor numerator, denominator, source statement and period, removal notes for three collinear factors, and data-availability tiers); factor exposure and quantile grouping (comp" }, "human": { "zh-CN": { "name": "柏讯 · 因子选股", "summary": "柏讯因子选股四件套：因子库与口径（52 因子九类清单、逐因子分子分母与报表期别、三个共线因子的剔除说明、数据可得性分档）、因子暴露与分层（走计算面，层数与极端值处理显式，IC 含 Pearson/Spearman，样本不足不作依据）、选股应用（多因子打分或条件筛选，产出候选 universe/目标权重，打分非预测）、无未来函数回测协议（财报用披露日而非报告期、universe 用当期成分、价格止于决策日、断言必须开启）。与自适应轮动套件的分工：轮动用价格类行业动量选组、本套件用基本面因子选股，两者可组合不重复；与个股投研套件的分工：那边做单只公司的深度报告，本套件做横截面的排序与筛选。因子值一律由计算面算出，不手算。回测为历史模拟，不构成投资建议。" }, "en": { "name": "Wire · Factor Selection", "summary": "Wire factor-selection four-skill suite: factor library and methodologies (a list of 52 factors in nine categories, with per-factor numerator, denominator, source statement and period, removal notes for three collinear factors, and data-availability tiers); factor exposure and quantile grouping (computed on the calculation surface, with explicit layer counts and outlier handling, IC incl. Pearson/Spearman, insufficient samples never used as a basis); stock-selection application (multi-factor scoring or conditional screening producing a candidate universe and target weights — scoring is not forecasting); and a no-look-ahead backtest protocol (financials dated by disclosure date, not reporting period; universe from current-period constituents; prices end at the decision date; assertions mandatory). Division of labor with the adaptive-rotation suite: rotation selects industry groups on price-based momentum, this suite selects stocks on fundamental factors — the two combine without overlap. With the single-stock research suite: that one produces in-depth reports on a single company, this one does cross-sectional ranking and screening. Factor values are always computed on the calculation surface, never by hand. Backtests are historical simulations and do not constitute investment advice." }, "es": { "name": "Wire · Selección por Factores", "summary": "Suite de cuatro habilidades de selección por factores de Wire: biblioteca de factores y criterios (lista de 52 factores en nueve categorías, con numerador, denominador, estado financiero y período de cada factor, notas de exclusión de tres factores colineales y niveles de disponibilidad de datos); exposición factorial y agrupación por capas (calculada en la superficie de cómputo, con número de capas y tratamiento de valores extremos explícitos, IC con Pearson/Spearman, y muestras insuficientes nunca como base); aplicación a la selección (puntuación multifactor o filtrado por condiciones que produce un universo candidato y ponderaciones objetivo; puntuar no es predecir); y protocolo de backtest sin sesgo de anticipación (estados financieros por fecha de revelación, no por período contable; universo con los constituyentes vigentes en cada fecha; precios hasta el día de decisión; aserciones obligatorias). Reparto de funciones con la suite de rotación adaptativa: la rotación elige grupos sectoriales por momentum de precios y esta suite selecciona acciones por factores fundamentales; combinables sin solaparse. Con la suite de investigación por empresa: aquella elabora informes en profundidad de una sola compañía, esta ordena y filtra en corte transversal. Los valores de los factores siempre los calcula la superficie de cómputo, nunca a mano. El backtest es una simulación histórica y no constituye asesoramiento de inversión." }, "fr": { "name": "Wire · Sélection par Facteurs", "summary": "Suite en quatre compétences de sélection par facteurs de Wire : bibliothèque de facteurs et méthodologies (liste de 52 facteurs en neuf catégories, avec numérateur, dénominateur, état financier et période de chaque facteur, notes d'exclusion de trois facteurs colinéaires et paliers de disponibilité des données) ; exposition factorielle et stratification (calculée sur la surface de calcul, avec nombre de strates et traitement des valeurs extrêmes explicites, IC avec Pearson/Spearman, échantillons insuffisants jamais retenus comme preuve) ; application à la sélection (score multi-facteurs ou filtrage conditionnel produisant un univers candidat et des pondérations cibles ; attribuer un score n'est pas prévoir) ; et protocole de backtest sans biais d'anticipation (états financiers datés à la date de publication et non à la période de reporting ; univers constitué des membres en vigueur ; prix arrêtés à la date de décision ; assertions obligatoires). Répartition des rôles avec la suite de rotation adaptative : la rotation choisit des groupes sectoriels sur le momentum des prix et celle-ci sélectionne des actions sur des facteurs fondamentaux — combinables sans chevauchement. Avec la suite de recherche par entreprise : celle-là produit des rapports approfondis sur une seule société, celle-ci trie et filtre en coupe transversale. Les valeurs de facteurs sont toujours calculées par la surface de calcul, jamais à la main. Le backtest est une simulation historique et ne constitue pas un conseil en investissement." }, "ja": { "name": "Wire · ファクター銘柄選定", "summary": "Wireファクター銘柄選定の4技能スイート：ファクターライブラリと算出基準（52ファクター9分類の一覧、ファクターごとの分子・分母・対象財務諸表と期別、共線3ファクターの除外理由、データ入手可能性のランク分け）、ファクターエクスポージャーと層別（計算サーフェスで実行、層数と外れ値処理を明示、ICはPearson/Spearman双方、サンプル不足は根拠にしない）、銘柄選定アプリケーション（マルチファクタースコアリングまたは条件スクリーニングで候補ユニバースと目標ウェイトを生成、スコアリングは予測ではない）、先読みなしバックテストプロトコル（財務データは報告期間ではなく開示日基準、ユニバースは当時点の構成銘柄、価格は意思決定日まで、アサーション必須）。アダプティブローテーションスイートとの役割分担：ローテーションは価格系の業種モメンタムでグループを選び、本スイートはファンダメンタルズ・ファクターで銘柄を選ぶ。両者は重複なく組み合わせ可能。個別銘柄リサーチスイートとの役割分担：そちらは単一企業のディープレポートを作り、本スイートは横断的なランキングとスクリーニングを担当。ファクター値はすべて計算サーフェスで算出し、手計算はしない。バックテストは履歴シミュレーションであり、投資助言を構成しない。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "factor-library", "description": "柏讯因子选股 · 因子库与口径。当用户问\"有哪些基本面因子可用\"\"这个因子怎么算\"\"为什么这两个因子不能同时用\"\"某个因子在 A 股跑不跑得动\"时使用。收录 FinRL-X 的 52 个基本面因子（九类），逐因子给出分子、分母、所用报表、期别、取值来源、数据可得性档位与 A 股适配说明，并写明三个共线因子的剔除理由。本技能只定义因子、不计算因子值、不选股、不做因子择时。" }, { "name": "factor-exposure", "description": "柏讯因子选股 · 因子暴露与分层。当用户要求\"算某只/某组标的的因子暴露\"\"按因子分层看收益差\"\"给我这个因子的 IC\"\"这个因子在横截面上有没有区分度\"时使用。暴露、分层、IC 与秩变换全部经 fd-quant 计算面完成，不在对话里手算；层数、极端值处理、样本门槛三处必须显式声明。产出：暴露表 + 分层收益表 + IC 数值与样本量。不做选股、不做因子择时。" }, { "name": "factor-screening", "description": "柏讯因子选股 · 选股应用。当用户要求\"按这些因子筛一批股票\"\"多因子打分选前 N 只\"\"把因子结论变成可回测的持仓\"时使用。把 factor-exposure 的暴露与 IC 结果变成候选 universe 与目标权重，权重或规则显式写出，筛选条件逐条可解释，缺失因子按声明规则处理。产出可直接进入权重模式回测的候选名单/权重。因子打分不是预测。" }, { "name": "factor-backtest-protocol", "description": "柏讯因子选股 · 无未来函数回测协议。当用户要把因子策略送进回测、要核对\"这次回测有没有偷看未来\"、要区分样本内与样本外、或要验某条未来数据会不会被拦住时使用。把四条点内纪律写成可核验的断言（财报按披露日、universe 用当期成分、价格止于决策日、断言必须开启），并用注入未来披露的夹具证明协议真的在拦。产出：回测口径卡 + 断言记录 + 夹具测试结果 + 样本内外分离的结论。不做因子择时、不给预期收益。" } ], "mcpServers": [ { "registryName": "fd-quant" }, { "registryName": "fd-cn-report" }, { "registryName": "fd-find-data-business-mcp" }, { "registryName": "fd-paper-trading" }, { "registryName": "fd-daas-mcp" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "BX0qjZe26KJU-waLQNVS-A", "slug": "pack-bx0qjze26kju-walqnvs-a", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "经济", "wire" ], "canonical": { "brief": "Wire event-driven four-skill set: event identification (three-way retrieval over announcement list / classification / full text, mapped into six event-type groups, each item with announcement id and date), event reading (read the body and extract elements — amount, ratio, horizon, party, effectivene" }, "human": { "zh-CN": { "name": "柏讯 · 事件驱动", "summary": "柏讯事件驱动四件套：事件识别（公告列表/分类/全文三路检索，六组事件类型映射，附公告标识与日期）、事件解读（读正文抽要素：金额/比例/期限/主体/生效条件，明确与需看条件二分，结论须引原文）、历史反应统计（多窗口收益分布 + 样本量 + 分位 + 必选的反例比例，明确是历史分布不是预测）、事件跟踪（按标的/类型筛选与统计，不含预测与推荐）。与柏讯舆情和投研套件的分工：舆情看文本倾向，投研做论点综合，本套件只处理结构化事件（公告/监管/业绩），不猜情绪也不给结论。核心纪律：不预测（不给涨跌预期、不说预计影响百分比）、正文支撑（标题与分类只是检索入口）、历史分布非预测（必须同时给反例比例）、预期已计入提示。公告是事实，解读是本套件的判断，两者分开呈现。不构成投资建议。" }, "en": { "name": "Wire · Event-Driven", "summary": "Wire event-driven four-skill set: event identification (three-way retrieval over announcement list / classification / full text, mapped into six event-type groups, each item with announcement id and date), event reading (read the body and extract elements — amount, ratio, horizon, party, effectiveness condition — split into explicit vs conditional, every conclusion citing the original text), historical reaction statistics (multi-window return distributions + sample size + quantiles + a mandatory counterexample ratio, explicitly a historical distribution, not a forecast), and event tracking (filter and tally by symbol or event type, no forecasts or recommendations). Division of labor with the Wire sentiment and equity research suites: sentiment reads text leaning, equity research synthesizes theses, this suite only handles structured events (announcements / regulatory / earnings) — it reads no mood and draws no conclusions. Core discipline: no prediction (no up/down expectations, no projected percentage impact), body-text support (title and category are only retrieval entries), historical distribution is not prediction (the counterexample ratio must accompany it), and the 'already priced in' reminder. Announcements are facts; the reading is this suite's judgment; the two are presented separately. Not investment advice." }, "es": { "name": "Wire · Basado en eventos", "summary": "Cuádruple de eventos de Wire: identificación de eventos (búsqueda triple por lista de anuncios / clasificación / texto completo, repartida en seis grupos de tipos de eventos, con identificador y fecha de cada anuncio), lectura de eventos (leer el cuerpo y extraer elementos — importe, proporción, plazo, sujeto, condición de entrada en vigor —, divididos entre explícitos y sujetos a condiciones, cada conclusión cita el texto original), estadística de reacciones históricas (distribuciones de retornos multi-ventana + tamaño muestral + cuantiles + ratio de contraejemplos obligatorio, distribución histórica y no predicción) y seguimiento de eventos (filtrar y contar por valor o tipo de evento, sin predicciones ni recomendaciones). Reparto con las suites de opinión pública e investigación de acciones de Wire: la opinión pública lee la inclinación del texto, la investigación sintetiza tesis, esta suite solo trata eventos estructurados (anuncios / regulatorios / resultados), sin adivinar ánimos ni dar conclusiones. Disciplina nuclear: no predecir (sin expectativas de subida o bajada, sin porcentajes de impacto proyectados), apoyo en el cuerpo del texto (título y clasificación son solo puertas de búsqueda), la distribución histórica no es predicción (el ratio de contraejemplos es obligatorio) y el aviso de «ya descontado». El anuncio es un hecho, la lectura es el juicio de esta suite; ambos se presentan por separado. No constituye asesoramiento de inversión." }, "fr": { "name": "Wire · Événementiel", "summary": "Quatuor d'événements Wire : identification des événements (recherche triple par liste d'annonces / classification / texte intégral, rangée en six familles de types d'événements, avec identifiant et date pour chaque annonce), lecture des événements (lire le corps et extraire les éléments — montant, proportion, échéance, partie, condition d'entrée en vigueur —, scindés entre explicites et sous conditions, chaque conclusion citant le texte original), statistiques des réactions historiques (distributions de rendements multi-fenêtres + taille d'échantillon + quantiles + ratio de contre-exemples obligatoire, une distribution historique et non une prévision) et suivi des événements (filtrer et compter par valeur ou type d'événement, sans prévisions ni recommandations). Répartition avec les suites Wire d'opinion publique et de recherche sur les actions : l'opinion publique lit l'inclinaison du texte, la recherche synthétise des thèses, cette suite ne traite que les événements structurés (annonces / réglementaires / résultats), sans deviner d'humeur ni conclure. Discipline centrale : pas de prédiction (aucune attente de hausse ou de baisse, aucun pourcentage d'impact projeté), étayage par le corps du texte (titre et classification ne sont que des portes d'entrée de recherche), la distribution historique n'est pas une prévision (le ratio de contre-exemples accompagne obligatoirement) et le rappel « déjà intégré aux cours ». L'annonce est un fait, la lecture est le jugement de cette suite ; les deux sont présentés séparément. Ne constitue pas un conseil en investissement." }, "ja": { "name": "Wire · イベントドリブン", "summary": "Wire イベントドリブンの 4 技能セット：イベント識別（公告リスト／分類／全文の 3 経路検索、6 系統のイベント型へのマッピング、公告 ID と日付付き）、イベント読解（本文を読んで要素——金額・比率・期限・主体・発効条件——を抽出し、「明確」と「条件次第」の二分、結論は原文を引用）、歴史的反応統計（複数窓のリターン分布＋サンプル数＋分位＋必須の反例比率、あくまで歴史分布であって予測ではない）、イベント追跡（銘柄・型別の絞り込みと集計、予測も推奨も含まない）。Wire センチメント・リサーチ両スイートとの分担：センチメントはテキストの傾向を見て、リサーチは論点を統合し、本スイートは構造化イベント（公告／規制／決算）だけを扱い、センチメントを推測せず結論も出さない。中核規律：予測しない（上昇・下落の期待も影響パーセントの見通しも出さない）、本文による支撑（タイトルと分類は検索の入口にすぎない）、歴史分布は予測ではない（反例比率を必ず併記）、「織り込み済み」の提示。公告は事実、読解は本スイートの判断であり、両者は分けて提示する。投資助言ではない。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "event-detection", "description": "柏讯事件驱动 · 事件识别与分类。当用户要求\"这家公司最近有什么公告/大事\"\"近三个月事件梳理\"\"哪些公司发了业绩预告\"\"把公告按事件类型分一下\"时使用。三路检索（公告列表 / 分类字典 / 全文检索），把公告归入六组事件类型（业绩类、融资类、股权变动、监管类、经营类、互动类），逐条附公告标识、披露日期、上游分类与归类依据。只做识别与分类：不读正文下结论、不判断利好利空、不做反应统计（分别交给 event-reading 与 event-history）。" }, { "name": "event-reading", "description": "柏讯事件驱动 · 事件解读。当用户要求\"这条公告到底说了什么\"\"这个定增/减持/业绩预告怎么看\"\"把公告关键条款摘出来\"时使用。读公告正文（get_filing），抽取金额、比例、期限、主体、生效条件五类要素，把结论分成\"正文要素直接支撑（明确）\"与\"需看条件\"两栏，每条结论引用原文并给出公告标识。硬约束：未读正文不结论，标题与分类只是检索入口。不做涨跌预测，不给买卖建议；对\"预期已计入\"只作事实呈现与一般性提示。" }, { "name": "event-history", "description": "柏讯事件驱动 · 历史反应统计。当用户要求\"这类事件历史上发生后表现如何\"\"业绩预增公告后 20 天一般怎么走\"\"帮我看看这个事件类型的历史分布\"时使用。对同类事件做多窗口（默认 5/10/20 个交易日）的历史反应统计：样本量、分位数、均值与离散度，并强制给出反例比例与样本门槛。全部输出为描述性历史分布，不是预测；样本不足不下结论；口径（按分类或按关键词）随结果一并写出。不提供涨跌预期、不给买卖建议。" }, { "name": "event-watch", "description": "柏讯事件驱动 · 事件跟踪。当用户要求\"帮我盯住这几家公司\"\"每天把公告按类型筛一遍\"\"按标的或事件类型给我一张事件跟踪表\"时使用。按标的与事件类型筛选公告流，输出事件清单、分类计数与期间对比（本周 vs 上周等），并明确标注检索口径与数据截至日期。不做预告、不做推荐、不预测影响；定时触发属宿主 cron 配置，本技能只提供被调用时的方法与输出形态，不实现调度。" } ], "mcpServers": [ { "registryName": "fd-cn-report" }, { "registryName": "fd-quant" }, { "registryName": "fd-daas-mcp" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "AlZiwYWeSfbs9hqRI4d-IA", "slug": "pack-alziwywesfbs9hqri4d-ia", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "经济", "金融", "wire" ], "canonical": { "brief": "Equity valuation and earnings-cycle analysis: DCF (five-year forecast + WACC + terminal value + sensitivity tables) + comparable companies + three-statement triangulation; earnings previews and post-earnings reviews (line-item localization and quality analysis). Financial data goes through the Wire " }, "human": { "zh-CN": { "name": "柏讯 · 估值套件", "summary": "股票估值与财报周期分析：DCF（五年预测 + WACC + 终值 + 敏感性表）+ 可比公司 + 三表联动三角验证；财报前瞻与复盘（科目定位与质量分析）。财务数据走柏讯服务端，股价与市值走本地数据平台通道。" }, "en": { "name": "Wire · Valuation Suite", "summary": "Equity valuation and earnings-cycle analysis: DCF (five-year forecast + WACC + terminal value + sensitivity tables) + comparable companies + three-statement triangulation; earnings previews and post-earnings reviews (line-item localization and quality analysis). Financial data goes through the Wire server side; share prices and market caps go through the local data platform channel." }, "es": { "name": "Wire · Suite de valoración", "summary": "Valoración de acciones y análisis del ciclo de resultados: DCF (pronóstico a cinco años + WACC + valor terminal + tablas de sensibilidad) + comparables + triangulación con los tres estados; anticipación y revisión de resultados (localización de partidas y análisis de calidad). Los datos financieros van por el lado servidor de Wire; precios y capitalización por el canal de la plataforma de datos local." }, "fr": { "name": "Wire · Suite de valorisation", "summary": "Valorisation d'actions et analyse du cycle des résultats : DCF (prévision à cinq ans + WACC + valeur terminale + tableaux de sensibilité) + comparables + triangulation des trois états ; anticipation et revue des résultats (localisation des postes et analyse de qualité). Les données financières passent par le côté serveur de Wire ; les cours et capitalisations par le canal de la plateforme de données locale." }, "ja": { "name": "Wire · バリュエーションスイート", "summary": "株式バリュエーションと決算サイクル分析：DCF（5 年予測＋WACC＋終価＋感応度表）＋類似会社＋三表連動のトライアンギュレーション。決算の事前展望と事後レビュー（科目の特定と質分析）。財務データは Wire サーバー側、株価と時価総額はローカルデータプラットフォームのチャネルを通す。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "equity-valuation", "description": "股票估值（DCF + 可比公司 + 三表联动）。当用户要求给某家 A 股/港股公司做内在价值或相对估值、算 WACC 与终值、跑敏感性与情景表、给目标价区间、做同业倍数对照与反推市场隐含预期时使用。财务数据走柏讯服务端通道 A，价格与市值走本地 DAAS 通道 B；通道 B 未连接时中止估值，不给目标价。" }, { "name": "earnings-cycle-analysis", "description": "财报周期分析（前瞻 + 复盘）。当用户要求对一家公司做财报前瞻（本次披露类型与时间、该盯哪些科目、超预期/低于预期可能从哪来）、做财报复盘（实际 vs 上年同期、差异由哪个科目驱动、盈利质量如何、对后续季度意味着什么），或梳理连续数期经营动能变化时使用。基于柏讯财报与公告数据，不做技术分析。触发词：财报前瞻、业绩预告、财报复盘、业绩点评、超预期、低于预期、盈利质量、某某公司季报/半年报/年报分析。" }, { "name": "stock-research-workflow", "description": "个股研究全流程工作流（数据-股票包入口技能）。当用户询问一只股票（A股或美股）的投资价值、要求个股分析、财报解读、估值判断或多空辩论时使用。驱动从行情基本面到估值结论的完整研究流程，调用行情工具、数据 MCP 与研究技能。" } ], "mcpServers": [ { "registryName": "fd-cn-report" }, { "registryName": "fd-daas-mcp" }, { "registryName": "fd-find-data-business-mcp" }, { "registryName": "fd-open-data-mcp" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "azkOwU4q6mmTmbw75Ow7yQ", "slug": "pack-azkowu4q6mmtmbw75ow7yq", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "投研", "个股", "wire" ], "canonical": { "brief": "Full-chain single-stock research: data acquisition and methodology, peers and competition, financial modeling, thesis synthesis, and 13-chapter report assembly. Five skills aligned with the five running roles of FinRobot production V2 (data / analysis / modeling / synthesis / report). Financials and" }, "human": { "zh-CN": { "name": "柏讯 · 个股投研套件", "summary": "个股深度投研：从取数与口径、同行与竞争、财务建模、论点综合到 13 章报告装配的完整链条。五技能对齐 FinRobot 生产版 V2 的五个在跑角色（数据/分析/建模/综合/报告）。财务与披露走柏讯服务端通道 A，价格与市值走本地通道 B，估值与统计数字一律由 fd-quant 计算面产出（不自行计算）。与「估值套件」（单点估值）、「财务底盘」套件的分工：本套件是**全链报告**，估值套件是**单点深挖**，财务底盘是**通用财务分析**。方法分歧时点位可缺（null），评级必发。" }, "en": { "name": "Wire · Equity Research Suite", "summary": "Full-chain single-stock research: data acquisition and methodology, peers and competition, financial modeling, thesis synthesis, and 13-chapter report assembly. Five skills aligned with the five running roles of FinRobot production V2 (data / analysis / modeling / synthesis / report). Financials and disclosures go through the Wire server-side channel A, prices and market caps through the local channel B, and all valuation and statistical figures are produced by the fd-quant compute plane (never computed in-house). Division of labor with the Valuation suite (single-point valuation) and the Financial Foundation suite: this suite is the full-chain report, the Valuation suite is the single-point deep dive, and the Financial Foundation suite is general-purpose financial analysis. When methods disagree, point values may be missing (null) but a rating is always issued." }, "es": { "name": "Wire · Suite de investigación de acciones", "summary": "Investigación de acciones de cadena completa: obtención de datos y criterios, comparables y competencia, modelización financiera, síntesis de tesis y ensamblaje del informe de 13 capítulos. Cinco skills alineados con los cinco roles en producción de FinRobot V2 (datos / análisis / modelización / síntesis / informe). Finanzas y disclosures van por el canal A del lado servidor de Wire, precios y capitalizaciones por el canal B local, y toda cifra de valoración o estadística la produce el plano de cálculo fd-quant (nunca se calcula por cuenta propia). Reparto con la Suite de valoración (valoración puntual) y la suite de base financiera: esta suite es el informe de cadena completa, la Suite de valoración es el análisis profundo de un punto y la base financiera es análisis financiero general. Cuando los métodos discrepan, el valor puntual puede faltar (null) pero el rating se emite siempre." }, "fr": { "name": "Wire · Suite de recherche sur les actions", "summary": "Recherche sur les actions en chaîne complète : acquisition de données et méthodologie, pairs et concurrence, modélisation financière, synthèse de thèse et assemblage du rapport en 13 chapitres. Cinq skills alignés sur les cinq rôles en production de FinRobot V2 (données / analyse / modélisation / synthèse / rapport). Les données financières et les disclosures passent par le canal A côté serveur de Wire, les prix et capitalisations par le canal B local, et tout chiffre de valorisation ou statistique est produit par le plan de calcul fd-quant (jamais calculé en interne). Répartition avec la Suite de valorisation (valorisation ponctuelle) et la suite de socle financier : cette suite fait le rapport de chaîne complète, la Suite de valorisation l'analyse ponctuelle approfondie, et le socle financier l'analyse financière générale. En cas de divergence des méthodes, la valeur ponctuelle peut manquer (null) mais la recommandation est toujours émise." }, "ja": { "name": "Wire · 個別銘柄リサーチスイート", "summary": "個別銘柄のフルチェーン型リサーチ：データ取得と算出基準、ピアと競合、財務モデリング、論点の統合、13 章レポートの組み立て。5 技能は FinRobot 生産版 V2 の稼働中 5 ロール（データ／分析／モデリング／統合／レポート）に対応。財務と開示は Wire サーバー側チャネル A、価格と時価総額はローカルチャネル B、バリュエーションと統計数値はすべて fd-quant 計算プレーンが生成（独自計算しない）。「バリュエーションスイート」（単点のバリュエーション）、「財務基盤」スイートとの分担：本スイートはフルチェーンのレポート、バリュエーションスイートは単点の深掘り、財務基盤は汎用の財務分析。手法が割れた場合はポイント値の欠落（null）は許容するが、レーティングは必ず発行する。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "equity-research-data", "description": "个股投研的取数与口径纪律层。当用户要求为某家 A 股/港股公司取财务科目、公告披露、行业财务指标或公司面板变量，或需要交代某个数字的来源工具、数据期数、币种与单位时使用。财务与披露一律走柏讯服务端通道 A，价格与市值一律走本地 DAAS 通道 B，两条通道不混用、不互相替代；取不到的数值如实声明缺口，不得填补。" }, { "name": "equity-research-peers", "description": "个股投研的同业与竞争层。当用户要求为某家 A 股/港股公司挑可比公司、给同业规模与增速带对照、做可比性说明或跑倍数对照表时使用。候选池用 GTA 公司面板按规模带与增速带筛选（通道 A），价格类输入与倍数只走本地 DAAS（通道 B）；同业之间的可比性差异必须逐项标注，不得抹平。" }, { "name": "equity-research-modeling", "description": "个股投研的估值建模与计算层。当用户要求给某家 A 股/港股公司做 DCF、DDM、可比倍数、Monte Carlo 或敏感性分析，或要求算基本面比率、跑红旗检测、把多种估值方法合成目标价时使用。全部算术一律调 fd-quant 确定性计算面（通道 Q），本技能只组织输入、标注来源、原样转述返回；计算面不可用时顶部明示\"估值部分未完成\"，绝不自算顶替。财务科目走柏讯服务端通道 A，价格与市值类输入走本地 DAAS 通道 B。" }, { "name": "equity-research-thesis", "description": "个股论点综合与方向判定（Rating 契约实例）。当用户要求给某只股票下方向结论、写投资论点、列催化剂与风险、给目标价，或把取数/同行/建模三段结果收敛成一个可入决策账本的判定时使用。产出 side / conviction / price_target（可 null）/ basis / evidence；方法分歧、价格通道不可用或输入不足时扣值不扣结论。财务与披露走柏讯服务端通道 A，本地库存读数走本地 DAAS 通道 B，估值与统计量走 fd-quant 计算面通道 Q。" }, { "name": "equity-research-report", "description": "个股投研报告装配（本套件入口技能）。当用户要求\"给某只股票出一份完整研究报告\"\"写一份研报\"\"跑一遍个股投研流程\"时从这里进入。按固定 13 章装配报告、执行六项叙述审计自检、保留缺数据章并写明原因、末尾给数据期数汇总行。串起取数与口径、同行比较、财务建模叙述、论点综合四段子流程，产出 13 章报告与一个 Rating 契约实例。" } ], "mcpServers": [ { "registryName": "fd-cn-report" }, { "registryName": "fd-daas-mcp" }, { "registryName": "fd-find-data-business-mcp" }, { "registryName": "fd-open-data-mcp" }, { "registryName": "fd-quant" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "wqpd5rPCFl3qIHg6oLDamg", "slug": "pack-wqpd5rpcfl3qihg6oldamg", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "复盘", "决策", "wire" ], "canonical": { "brief": "Turns 'the call made at the time' into an auditable history: decision logging (append-only, idempotent, revisions leave traces) → settlement at expiry against a benchmark for realized alpha → distilled reflection (injected only into the decision-maker, never broadcast) → portfolio check-up and consi" }, "human": { "zh-CN": { "name": "柏讯 · 决策复盘套件", "summary": "把「当时的判定」变成可审计的历史：决策落账（追加式、幂等、修订留痕）→ 到期按基准结算真实 alpha → 提炼反思（只注入决策者，不广播）→ 持仓体检与最近判定的一致性核查。以账本为准，不美化历史；样本不足不下结论。与决策类 pack 的分工：决策 pack 产出 Rating/TargetWeights 实例，本套件负责统一落账与复盘（语义相同，不重复实现）。" }, "en": { "name": "Wire · Decision Review Suite", "summary": "Turns 'the call made at the time' into an auditable history: decision logging (append-only, idempotent, revisions leave traces) → settlement at expiry against a benchmark for realized alpha → distilled reflection (injected only into the decision-maker, never broadcast) → portfolio check-up and consistency check against the most recent calls. The ledger is the source of truth and history is never embellished; no conclusions on insufficient samples. Division of labor with decision packs: decision packs produce Rating/TargetWeights instances, this suite handles unified logging and review (same semantics, no duplicate implementation)." }, "es": { "name": "Wire · Suite de revisión de decisiones", "summary": "Convierte «el juicio de aquel momento» en un historial auditable: registro de decisiones (solo aditivo, idempotente, las revisiones dejan rastro) → liquidación al vencimiento frente al índice de referencia para el alfa real → destilación de reflexiones (se inyectan solo al decisor, nunca se difunden) → chequeo de cartera y verificación de coherencia con los juicios más recientes. El registro manda y el historial no se embellece; sin conclusiones con muestras insuficientes. Reparto con los packs de decisión: estos producen instancias Rating/TargetWeights y esta suite se encarga del registro unificado y de la revisión (misma semántica, sin implementación duplicada)." }, "fr": { "name": "Wire · Suite de revue des décisions", "summary": "Transforme « le jugement porté sur le moment » en un historique auditable : enregistrement des décisions (append-only, idempotent, les révisions laissent une trace) → règlement à l'échéance contre l'indice de référence pour l'alpha réel → distillation de réflexions (injectées uniquement au décideur, jamais diffusées) → bilan de portefeuille et contrôle de cohérence avec les jugements les plus récents. Le registre fait foi et l'historique n'est jamais embelli ; pas de conclusion sur échantillon insuffisant. Répartition avec les packs de décision : ceux-ci produisent des instances Rating/TargetWeights, cette suite assure l'enregistrement unifié et la revue (même sémantique, sans implémentation dupliquée)." }, "ja": { "name": "Wire · 意思決定レビュースイート", "summary": "「その時の判定」を監査可能な履歴に変える：意思決定の記帳（追記式・冪等・改訂は痕跡を残す）→ 期限到来時にベンチマーク対比で実アルファを決済 → 振り返りの精製（意思決定者にのみ注入し、配信しない）→ ポートフォリオ診断と直近判定との整合性チェック。台帳を正とし、履歴を美化しない。サンプル不足では結論を出さない。意思決定系 pack との分担：意思決定 pack が Rating／TargetWeights インスタンスを生成し、本スイートは統一的な記帳とレビューを担う（意味論は同じ、実装の重複なし）。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "decision-review-log", "description": "柏讯决策落账纪律（复盘套件入口）。当用户要求把一条判定（Rating）或目标权重（TargetWeights）记入决策账本、要求\"再记一次\"、要求修改或\"更正\"已落账的判定、或需要查回某标的的历史判定与教训时使用。账本是追加式的：幂等键 (caller, as_of, symbol, pack, skill) 拒绝重复提交并返回既有 decision_id；记录按 caller 隔离，只看到自己的判定，修订产生新记录并以 revises 指向被修订记录，历史判定永不被就地改写。" }, { "name": "decision-review-settlement", "description": "柏讯到期结算复盘。当用户要求结算某条已落账的判定、想知道某判定\"后来怎么样了\"、要求汇总待结算记录、或要看原始收益与相对基准 alpha 时使用。流程为待结算汇总 → 取结算窗序列（价格走本地 DAAS 通道 B）→ 触发账本结算 → 三段式输出（判定理由回顾 + 结算结果 + 对照评述）。持有期未满不强行结算；alpha 为负不等于判断错误。" }, { "name": "decision-review-reflection", "description": "柏讯决策反思提炼。当用户要求为一条已结算（或已落账）的判定写复盘反思、要求\"总结这次教训\"、或询问为什么反思不该出现在其他角色的输出里时使用。反思固定 2–4 句，针对决策本身而非市场走势，区分\"判断错\"与\"运气差\"，并含可执行的改进项；反思由账本存储、只注入最终决策者的上下文，不得广播给其他角色，缺失反思不阻塞结算。" }, { "name": "decision-review-portfolio", "description": "柏讯持仓体检。当用户要求体检当前模拟账户持仓、看集中度（HHI/前 N 权重）、期间回撤、行业暴露，或核查\"持仓是否偏离了最近一次判定\"时使用。持仓与账户取自柏讯模拟交易服务（paper-only，模拟盘，无真实资金），统计量一律交给 fd-quant 计算面，缺失数据如实标注，结论为研究参考、不含买卖建议。" } ], "mcpServers": [ { "registryName": "fd-cn-report" }, { "registryName": "fd-daas-mcp" }, { "registryName": "fd-paper-trading" }, { "registryName": "fd-quant" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "8edSwnv3lIXcuSAkzS9vvA", "slug": "pack-8edswnv3lixcusakzs9vva", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "数据治理", "台账", "wire" ], "canonical": { "brief": "Wire data governance suite: read the data collected through forms, correct one record, undo a correction, review correction history, and export CSV. Written corrections must carry a source marker (source) and never fill values on the user's behalf; undoing only sets a rollback marker and the origina" }, "human": { "zh-CN": { "name": "柏讯 · 数据治理", "summary": "柏讯数据治理套件：读表单收上来的数据、改错一条记录、撤销改错、查改错历史、导出 CSV。写修正必须带来源标记（source）且不得代填；撤销只置回滚标记，原始行永不改写；导出带 truncated 字段，被行数闸截断时如实告知。失败原样转述，禁止以空列表冒充无数据。" }, "en": { "name": "Wire · Data Governance", "summary": "Wire data governance suite: read the data collected through forms, correct one record, undo a correction, review correction history, and export CSV. Written corrections must carry a source marker (source) and never fill values on the user's behalf; undoing only sets a rollback marker and the original row is never rewritten; exports carry a truncated field, and truncation by the row cap is reported as-is rather than passed off as complete. Failures are relayed verbatim; an empty list must never pose as 'no data'." }, "es": { "name": "Wire · Gobernanza de datos", "summary": "Suite de gobernanza de datos de Wire: leer los datos recogidos mediante formularios, corregir un registro, deshacer una corrección, consultar el historial de correcciones y exportar CSV. Toda corrección escrita debe llevar marcador de origen (source) y nunca rellena valores por el usuario; deshacer solo activa una marca de reversión y la fila original jamás se reescribe; la exportación incluye el campo truncated y si el tope de filas cortó el resultado se informa tal cual, sin presentarlo como completo. Los fallos se transmiten literalmente; queda prohibido hacer pasar una lista vacía por «sin datos»." }, "fr": { "name": "Wire · Gouvernance des données", "summary": "Suite de gouvernance des données Wire : lire les données collectées via les formulaires, corriger un enregistrement, annuler une correction, consulter l'historique des corrections et exporter en CSV. Toute correction écrite doit porter un marqueur de source (source) et ne renseigne jamais de valeur à la place de l'utilisateur ; l'annulation se contente de poser un marqueur de rollback et la ligne d'origine n'est jamais réécrite ; l'export comporte un champ truncated, et une troncature par le garde-fou de lignes se signale telle quelle, sans se présenter comme complète. Les échecs sont relayés mot pour mot ; faire passer une liste vide pour « aucune donnée » est interdit." }, "ja": { "name": "Wire · データガバナンス", "summary": "Wire データガバナンススイート：フォームで収集したデータの読み取り、1 件のレコード修正、修正の取り消し、修正履歴の照会、CSV エクスポート。修正の書き込みにはソースマーク（source）が必須で、代わりに値を埋めない。取り消しはロールバックマークを立てるだけで、元の行は決して書き換えない。エクスポートには truncated フィールドがあり、行数ゲートで切り詰められた場合は全量ではないとありのままに伝える。ツールの失敗は原文のまま伝達し、空リストで「データなし」を装うことは禁止。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "wire-data-steward", "description": "柏讯数据治理。当用户要读表单收上来的数据、改错一条记录、撤销一次改错、查改错历史或导出 CSV 时使用。写修正必须带来源标记（source），不得代填；撤销只置回滚标记、原始行永不改写；导出结果带 truncated 字段，true 表示被上游行数闸截断而非全量——必须如实转述。任何工具失败原样转述，禁止以空列表冒充无数据。" } ], "mcpServers": [ { "registryName": "wire-build-mcp" } ] } }, { "id": "SWPvacoFTftfVngMMFotwQ", "slug": "pack-swpvacoftftfvngmmfotwq", "category": "pack", "marketCategory": "wire", "tags": [ "柏讯", "产物", "看板", "表单", "wire" ], "canonical": { "brief": "Wire artifact build suite: turn data into things visible in the console through conversation — build data-entry forms, build single-record scan pages, edit artifact specs, publish and archive. Drafts first, publishing separate; column names come from facts fetched first, never invented; upstream fai" }, "human": { "zh-CN": { "name": "柏讯 · 产物构建", "summary": "柏讯产物构建套件：用对话把数据变成控制台里看得见的东西——建数据录入表单、建单记录扫码页、改产物 spec、发布与归档。草稿优先、发布独立；列名先取事实，禁止编造；失败原样转述。附带一支三岗团队模板（巡线岗主动、内容岗与源岗被动），把「持续迭代内容」编成可安装的组织形态。" }, "en": { "name": "Wire · Artifact Build", "summary": "Wire artifact build suite: turn data into things visible in the console through conversation — build data-entry forms, build single-record scan pages, edit artifact specs, publish and archive. Drafts first, publishing separate; column names come from facts fetched first, never invented; upstream failures are relayed verbatim. Ships with a three-post team template (proactive line patrol, reactive content and source posts) that turns 'continuously iterating content' into an installable organizational form." }, "es": { "name": "Wire · Construcción de artefactos", "summary": "Suite Wire de construcción de artefactos: convertir datos en cosas visibles en la consola mediante conversación — crear formularios de captura de datos, crear páginas de escaneo de registro único, editar el spec de un artefacto, publicar y archivar. Primero borrador, publicación aparte; los nombres de columna se toman primero de los hechos, prohibido inventar; los fallos aguas arriba se transmiten tal cual. Incluye una plantilla de equipo de tres puestos (ronda de línea proactiva, puestos de contenido y de fuente reactivos) que convierte «iterar contenido continuamente» en una forma organizativa instalable." }, "fr": { "name": "Wire · Construction d'artefacts", "summary": "Suite Wire de construction d'artefacts : transformer des données en éléments visibles dans la console par la conversation — créer des formulaires de saisie de données, créer des pages de scan à enregistrement unique, modifier le spec d'un artefact, publier et archiver. Le brouillon d'abord, la publication à part ; les noms de colonnes proviennent d'abord des faits, interdiction d'inventer ; les échecs amont sont retranscrits tels quels. Livrée avec un modèle d'équipe à trois postes (ronde de ligne proactive, postes contenu et source réactifs) qui transforme « itérer continûment le contenu » en une forme organisationnelle installable." }, "ja": { "name": "Wire · 成果物構築", "summary": "Wire 成果物構築スイート：会話を通じてデータをコンソールで見える形に変える——データ入力フォームの作成、単一レコードのスキャンページ作成、成果物 spec の変更、公開とアーカイブ。下書き優先、公開は独立。列名は先に事実を取得し、捏造禁止。上流の失敗はそのまま伝える。3 役のチームテンプレート（巡回役が能動、コンテンツ役とソース役は受動）を同梱し、「コンテンツの継続的改善」をインストール可能な組織形態に編成する。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "wire-artifact-build", "description": "柏讯产物构建。当用户要用对话把数据变成柏讯里看得见的东西——建看板（dashboard）、建正文文档（doc）、建数据录入表单（form）、建单记录扫码页（record-view）、给已有产物改 spec、发布或归档产物——时使用。所有写入一律落草稿，发布必须是独立的第二次显式调用；spec 引用的列名必须先用工具取真实列清单核对，不得凭记忆编造；上游拒绝时原样转述错误，不伪造成功。" }, { "name": "fleet-signal-triage", "description": "巡线信号五类归并与委派三要素：无信号是合法结论，写入工具零调用。" }, { "name": "fleet-artifact-draft-discipline", "description": "产物草稿纪律：schema 预检、数值来源标注、草稿与发布分离、source_ref 身份三元组。" } ], "mcpServers": [ { "registryName": "wire-build-mcp" } ] } }, { "id": "noL2XKaTZP2kiERrC_gtKA", "slug": "pack-nol2xkatzp2kierrc-gtka", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "轮动", "策略", "wire" ], "canonical": { "brief": "Adaptive industry rotation: from market-regime classification (slow trend/drawdown/volatility Z plus a fast shock check) to group-strength ranking (industry indices synthesized from SW Level-1 constituents), intra-group ranking (residual momentum + robust Z + an exception framework), risk-budgeted p" }, "human": { "zh-CN": { "name": "柏讯 · 自适应轮动套件", "summary": "自适应行业轮动：从市场状态判定（慢速趋势/回撤/波动率 Z + 快速冲击）到组强度排序（申万 L1 成分股合成的行业指数）、组内排序（残差动量 + 稳健 Z + 例外框架）、风险预算建仓（产出 TargetWeights 契约实例）与闭环运行（权重模式回测 + 模拟盘 + 快照解读）。对齐 FinRL-X 的七个策略模块，但输入已按 A 股适配：波动率用已实现波动率（QVIX/iVIX 在柏讯可及环境不可达），组用申万成分股合成（官方指数序列未采），参数按实测分位重标（原 Z≥3.0 → 采用 |Z|≥2.0，因实测 3.0 只占 0.47%）。价格走通道 B（本地 DAAS，需先播种 A 股），统计一律由 fd-quant 计算面产出。与「策略工坊」的分工：工坊给通用纪律（权重语义、回测解读、风控归属），本套件给**具体策略与参数**（选哪些组、组内选谁、预算怎么分）；督导不重复实现。终产物是可执行权重与运行记录，不是买入建议。" }, "en": { "name": "Wire · Adaptive Rotation Suite", "summary": "Adaptive industry rotation: from market-regime classification (slow trend/drawdown/volatility Z plus a fast shock check) to group-strength ranking (industry indices synthesized from SW Level-1 constituents), intra-group ranking (residual momentum + robust Z + an exception framework), risk-budgeted position building (emitting TargetWeights contract instances) and closed-loop operation (weight-mode backtests + paper trading + snapshot reading). Aligned with FinRL-X's seven strategy modules, but inputs are adapted to A-shares: volatility uses realized volatility (QVIX/iVIX are unreachable from Wire's reachable environment), groups are synthesized from SW constituents (official index series not sampled), and parameters are re-calibrated to measured quantiles (original Z≥3.0 → adopted |Z|≥2.0, since measured 3.0 covers only 0.47%). Prices go through channel B (local DAAS; A-shares must be seeded first); statistics are always produced by the fd-quant compute face. Division of labor with the strategy-lab suite: the lab provides general discipline (weight semantics, backtest interpretation, risk-control ownership), this suite provides the concrete strategy and parameters (which groups, which names within a group, how the budget splits); supervision does not re-implement. The end product is executable weights and run records, not buy advice." }, "es": { "name": "Wire · Suite de rotación adaptativa", "summary": "Rotación sectorial adaptativa: desde la clasificación de régimen de mercado (tendencia lenta/drawdown/volatilidad Z más un control rápido de choques) hasta el ranking de fuerza de grupos (índices sectoriales sintetizados a partir de los constituyentes SW L1), el ranking dentro del grupo (momentum residual + Z robusta + un marco de excepciones), la construcción de posiciones con presupuesto de riesgo (emite instancias del contrato TargetWeights) y la operación en bucle cerrado (backtests en modo pesos + trading simulado + lectura de instantáneas). Alineada con los siete módulos de estrategia de FinRL-X, pero con entradas adaptadas a acciones A: la volatilidad usa la volatilidad realizada (QVIX/iVIX son inalcanzables desde el entorno accesible de Wire), los grupos se sintetizan con constituyentes SW (las series de índices oficiales no se muestrean) y los parámetros se recalibran con cuantiles medidos (Z≥3.0 original → se adopta |Z|≥2.0, pues 3.0 medido solo cubre el 0,47 %). Los precios van por el canal B (DAAS local; primero hay que sembrar las acciones A); la estadística la produce siempre el plano de cálculo fd-quant. Reparto con la suite «laboratorio de estrategias»: el laboratorio aporta disciplina general (semántica de pesos, lectura de backtests, titularidad del control de riesgo), esta suite aporta la estrategia y los parámetros concretos (qué grupos, qué nombres dentro del grupo, cómo repartir el presupuesto); la supervisión no reimplementa. El producto final son pesos ejecutables y registros de operación, no una recomendación de compra." }, "fr": { "name": "Wire · Suite de rotation adaptative", "summary": "Rotation sectorielle adaptative : de la classification de régime de marché (tendance lente/drawdown/volatilité Z plus un contrôle rapide de choc) au classement de force des groupes (indices sectoriels synthétisés à partir des constituants SW L1), au classement intra-groupe (momentum résiduel + Z robuste + un cadre d'exceptions), à la construction de positions à budget de risque (émet des instances du contrat TargetWeights) et à l'exploitation en boucle fermée (backtests en mode poids + paper trading + lecture d'instantanés). Alignée sur les sept modules de stratégie de FinRL-X, mais avec des entrées adaptées aux actions A : la volatilité utilise la volatilité réalisée (QVIX/iVIX sont inaccessibles depuis l'environnement joignable de Wire), les groupes sont synthétisés à partir des constituants SW (les séries d'indices officiels ne sont pas échantillonnées) et les paramètres sont recalibrés sur les quantiles mesurés (Z≥3.0 d'origine → |Z|≥2.0 retenu, car 3.0 mesuré ne couvre que 0,47 %). Les prix passent par le canal B (DAAS local ; les actions A doivent être chargées d'abord) ; la statistique est toujours produite par le plan de calcul fd-quant. Répartition avec la suite « laboratoire de stratégies » : le laboratoire fournit la discipline générale (sémantique des poids, lecture des backtests, imputation du contrôle des risques), cette suite fournit la stratégie et les paramètres concrets (quels groupes, quels noms dans le groupe, comment répartir le budget) ; la supervision ne réimplémente pas. Le produit final est des poids exécutables et des enregistrements d'exécution, pas un conseil d'achat." }, "ja": { "name": "Wire · 適応型ローテーションスイート", "summary": "適応型業界ローテーション：市場レジーム判定（低速トレンド／ドローダウン／ボラティリティ Z＋高速ショック検知）から、グループ強度ランキング（SW L1 構成銘柄から合成した業界指数）、グループ内ランキング（残差モメンタム＋ロバスト Z＋例外フレーム）、リスク予算による建玉（TargetWeights 契約インスタンスを出力）、クローズドループ運用（ウェイトモードのバックテスト＋ペーパートレード＋スナップショット読解）へと進む。FinRL-X の 7 戦略モジュールに対応するが、入力は A 株向けに調整済み：ボラティリティは実現ボラティリティを使用（QVIX/iVIX は Wire から到達可能な環境では利用不可）、グループは SW 構成銘柄から合成（公式指数系列は未取得）、パラメータは実測分位で再調整（元の Z≥3.0 → 採用 |Z|≥2.0。実測では 3.0 は 0.47% しか覆わないため）。価格はチャネル B（ローカル DAAS、A 株データの事前取り込みが必要）、統計はすべて fd-quant 計算面が生成。「戦略ラボ」スイートとの分担：ラボは汎用規律（ウェイトの意味論、バックテスト読解、リスク管理の帰属）を提供し、本スイートは具体的な戦略とパラメータ（どのグループを選ぶか、グループ内でどの銘柄を選ぶか、予算をどう分けるか）を提供する。監督役が再実装することはない。最終成果物は実行可能なウェイトと運用記録であって、買い推奨ではない。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "rotation-regime", "description": "柏讯自适应轮动 · 市场状态判定（套件入口）。当用户要求判断当前 A 股市场处于进攻/中性/防守状态、要求给出轮动的择时前提、或问\"现在该不该加仓/降仓\"时使用。产出：带证据的 risk_on / neutral / risk_off 判定 + 波动率输入说明 + 参数重标表。本技能只做判定，不选股不建仓——排序与建仓交给 rotation-ranking / rotation-portfolio。" }, { "name": "rotation-ranking", "description": "柏讯自适应轮动 · 组与组内排序。当用户要求\"哪些行业值得配\"\"组内选哪几只\"\"为什么这个组落选\"时使用。组 = 申万 L1（31 类）成分股合成的行业指数；组强度用信息比排序，组内用残差动量 + 稳健 Z。产出：组排序表 + 组内排序表 + 落选原因。不建仓（交给 rotation-portfolio）。" }, { "name": "rotation-portfolio", "description": "柏讯自适应轮动 · 风险预算建仓与风控参数。当用户要求\"按排序结果给个仓位方案\"\"组之间怎么分配\"\"止损怎么设\"时使用。产出：TargetWeights 契约实例（组权重 → 资产权重）+ 风险预算依据 + 风控参数（并明确止损须在策略内表达、引擎不自动执行）。" }, { "name": "rotation-run", "description": "柏讯自适应轮动 · 闭环运行与解读。当用户要\"把这份权重跑一下\"\"回测结果怎么看\"\"模拟盘为什么没成交/权重为什么偏了\"时使用。产出：权重模式回测结果（含成本与基准）+ 权重快照解读 + 偏离归因。终产物是可执行权重与运行记录，不是一段买入建议。" } ], "mcpServers": [ { "registryName": "fd-quant" }, { "registryName": "fd-paper-trading" }, { "registryName": "fd-daas-mcp" }, { "registryName": "fd-cn-report" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "tXQaAuRY2F-Kog-rb21bCw", "slug": "pack-txqaaury2f-kog-rb21bcw", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "经济", "金融", "wire" ], "canonical": { "brief": "Wire Info Shock four-skill set: shock identification (source health first, then cursor pull plus retrieval as two complementary paths, initial triage by three source types and seven shock-type groups, each item with item_id, publish time and body status), shock reading (read the body and extract fiv" }, "human": { "zh-CN": { "name": "柏讯 · 信息冲击", "summary": "柏讯信息冲击四件套：冲击识别（先查来源健康度，游标拉取+检索补漏两路，按来源三型与七组冲击类型初筛，逐条附 item_id、发布时间与正文状态）、冲击解读（读正文抽金额/比例/期限/主体/生效条件五要素，结论分明确与需看条件两栏，每条结论附证据块：item_id + as_of + 经 locate_evidence 核对的原文逐字片段；强断言缺可定位证据时只作「据标题/摘要，正文未核实」弱表述）、影响窗口统计（前窗 5 日基准 + 后窗 5/10/20 日并列，样本量、分位、离散度与必选的反例比例随表输出，样本不足整表标注「不作依据」）、信息流跟踪（来源与类型计数、期间对比、健康度水位，滞后与不可用前置「信息可能不完整」）。与事件驱动套件的分工：事件驱动走 fd-cn-report 公告面，本套件走 info-stream 统一信息条目（公告+部委+互动），以 item_id 为引用锚点、以逐字证据为结论门槛。核心纪律：健康度前置（引用前先查 stream_health，lag 超阈值声明信息可能不完整，空结果不得写成没有事件）、不预测（不给涨跌预期）、正文支撑（标题只是检索入口）、历史分布非预测（样本量与反例比例强制随表）。不构成投资建议。" }, "en": { "name": "Wire · Info Shock", "summary": "Wire Info Shock four-skill set: shock identification (source health first, then cursor pull plus retrieval as two complementary paths, initial triage by three source types and seven shock-type groups, each item with item_id, publish time and body status), shock reading (read the body and extract five element classes — amount, ratio, horizon, party, effectiveness condition — with conclusions split into 'explicit' vs 'depends on conditions'; every conclusion carries an evidence block: item_id + as_of + a verbatim quote verified through locate_evidence; strong assertions without locatable evidence are downgraded to 'per title/abstract, body unverified'), impact-window statistics (front 5-day baseline vs back 5/10/20 trading days side by side, sample size, quantiles, dispersion and a mandatory counterexample ratio accompany every table; insufficient samples mark the whole table 'not to be relied on'), and stream tracking (source and type counts, period comparison, health watermark; lags and outages are prefixed 'information may be incomplete'). Division of labor with the event-driven suite: event-driven works the fd-cn-report announcement face, this suite works info-stream unified items (announcements + ministries + interactions), with item_id as the citation anchor and verbatim evidence as the bar for conclusions. Core discipline: health first (check stream_health before citing; if lag exceeds the threshold, declare 'information may be incomplete'; never report an empty result as 'no events'), no prediction (no up/down expectations), body-text support (a title is only a search entry), historical distribution is not forecast (sample size and counterexample ratio are mandatory). Not investment advice." }, "es": { "name": "Wire · Choque informativo", "summary": "Suite Wire de choque informativo en cuatro habilidades: identificación de choques (primero la salud de la fuente, luego extracción por cursor más búsqueda de refuerzo como dos vías complementarias, preselección por tres tipos de fuente y siete grupos de tipos de choque, cada ítem con item_id, hora de publicación y estado del cuerpo), lectura del choque (leer el cuerpo y extraer cinco clases de elementos —importe, ratio, plazo, parte, condición de vigencia—, con conclusiones divididas en «explícito» y «depende de condiciones»; cada conclusión lleva un bloque de evidencia: item_id + as_of + cita literal verificada con locate_evidence; las afirmaciones fuertes sin evidencia localizable se degradan a «según título/resumen, cuerpo sin verificar»), estadística de ventanas de impacto (base de 5 días previos frente a 5/10/20 días bursátiles posteriores en paralelo; tamaño muestral, cuantiles, dispersión y una proporción de contraejemplos obligatoria acompañan cada tabla; con muestra insuficiente, la tabla entera se marca «no sirve como fundamento»), y seguimiento del flujo (conteos por fuente y tipo, comparación entre periodos, marca de agua de salud; retrasos e indisponibilidades se anuncian con «la información puede estar incompleta»). Reparto con la suite basada en eventos: aquella trabaja la cara de anuncios de fd-cn-report, esta los ítems unificados de info-stream (anuncios + ministerios + interacción), con item_id como ancla de citación y la evidencia literal como umbral de conclusión. Disciplina central: salud primero (consultar stream_health antes de citar; si el retraso supera el umbral, declarar «la información puede estar incompleta»; nunca convertir un resultado vacío en «no hubo eventos»), sin predicción (sin expectativas de subida o bajada), soporte en el cuerpo (el título es solo puerta de búsqueda), la distribución histórica no es predicción (tamaño muestral y contraejemplos obligatorios). No es asesoramiento de inversión." }, "fr": { "name": "Wire · Choc d'information", "summary": "Suite Wire « choc d'information » en quatre compétences : identification des chocs (santé de la source d'abord, puis tirage par curseur et recherche de renfort comme deux voies complémentaires, pré-triage par trois types de source et sept familles de types de choc, chaque item avec item_id, heure de publication et statut du corps), lecture du choc (lire le corps et extraire cinq classes d'éléments — montant, ratio, horizon, partie, condition d'entrée en vigueur —, conclusions réparties en « explicite » et « dépend des conditions » ; chaque conclusion porte un bloc de preuve : item_id + as_of + citation verbatim vérifiée via locate_evidence ; les affirmations fortes sans preuve localisable sont rétrogradées en « d'après le titre/le résumé, corps non vérifié »), statistique de fenêtres d'impact (référence de 5 jours avant face à 5/10/20 jours de bourse après en parallèle ; taille d'échantillon, quantiles, dispersion et ratio de contre-exemples obligatoire accompagnent chaque tableau ; échantillon insuffisant → tableau entier marqué « ne sert pas de fondement »), et suivi du flux (comptages par source et type, comparaison entre périodes, filigrane de santé ; retards et indisponibilités annoncés par « l'information peut être incomplète »). Répartition avec la suite événementielle : celle-ci travaille la face annonces de fd-cn-report, celle-là les items unifiés d'info-stream (annonces + ministères + interactions), item_id comme ancre de citation et preuve verbatim comme seuil de conclusion. Discipline centrale : santé d'abord (consulter stream_health avant de citer ; au-delà du seuil de retard, déclarer « l'information peut être incomplète » ; jamais transformer un résultat vide en « aucun événement »), pas de prévision (pas d'attente de hausse ou de baisse), support par le corps (le titre n'est qu'une porte de recherche), la distribution historique n'est pas une prévision (taille d'échantillon et contre-exemples obligatoires). Ne constitue pas un conseil en investissement." }, "ja": { "name": "Wire · 情報ショック", "summary": "Wire 情報ショック 4 技能セット：ショック識別（まずソース健全性を確認し、カーソル取得と検索補完の 2 経路で情報フローを引き、3 系統のソース型と 7 系統のショック型で一次ふるい分け。各項目に item_id・公開時刻・本文状態を付す）、ショック読解（本文を読んで金額・比率・期限・主体・発効条件の 5 要素を抽出し、結論を「明確」と「条件次第」の二欄に分ける。各結論に根拠ブロック——item_id＋as_of＋locate_evidence で照合した原文の逐語引用——を添え、位置づけ可能な根拠のない強い断定は「見出し・要約によるもので本文は未確認」へ弱表現に落とす）、影響窓統計（前窓 5 営業日の基準＋後窓 5／10／20 営業日を並列、サンプル数・分位・ばらつきと必須の反例比率を表に添付し、サンプル不足の表は全体に「判断材料としない」を標記）、情報フロー追跡（ソース別・型別の計数、期間対比、健全性ウォーターマーク。遅延と障害には「情報が不完全な可能性」を前置き）。イベントドリブン・スイートとの分担：あちらは fd-cn-report の公告面、こちらは info-stream 統合項目（公告＋省庁＋インタラクション）を扱い、item_id を引用アンカー、逐語根拠を結論の関門とする。中核規律：健全性の前置（引用前に stream_health を確認、閾値超の遅延では「情報が不完全な可能性」を明示、空結果を「イベントなし」と書かない）、予測しない（上昇・下落の期待を出さない）、本文による支撑（見出しは検索の入口にすぎない）、歴史分布は予測ではない（サンプル数と反例比率は必須）。投資助言ではない。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "shock-detection", "description": "柏讯信息冲击 · 冲击识别。当用户要求\"最近有什么新公告/新信息\"\"把今天的信息流筛一遍\"\"找出可能构成信息冲击的条目\"时使用。先查来源健康度，再按游标拉信息流（list_since），按来源三型（公告/部委/互动）与七组冲击类型做初筛，逐条附 item_id、发布时间、正文状态与初筛依据。只做识别与初筛：不读正文下结论（交 shock-reading）、不做窗口统计（交 shock-window）、不预测影响。" }, { "name": "shock-reading", "description": "柏讯信息冲击 · 冲击解读。当用户要求\"这条公告/这条信息说了什么\"\"把关键条款摘出来\"\"这个消息的具体金额和条件\"时使用。读条目正文，抽取金额/比例/期限/主体/生效条件五类要素，结论分\"明确（正文要素直接支撑）\"与\"需看条件\"两栏；每条结论附证据块：item_id、as_of（发布时间）与经 locate_evidence 核对的原文逐字片段。硬约束：未读正文不结论；强断言缺可定位证据时只能弱表述\"据标题/摘要，正文未核实\"。不预测涨跌。" }, { "name": "shock-window", "description": "柏讯信息冲击 · 影响窗口统计。当用户要求\"这类公告发布前后一般怎么走\"\"这条信息的历史前后窗口表现\"\"前后窗口指标偏离多大\"时使用。对同类信息冲击做多窗口（默认前窗 5 日基准 + 后窗 5/10/20 个交易日）的历史统计：样本量、分位、离散度与必选的反例比例；样本不足时整表标注\"不作依据\"。全部输出为描述性历史分布，不是预测；价格只走通道 B，计算只走通道 Q。" }, { "name": "shock-watch", "description": "柏讯信息冲击 · 信息流跟踪。当用户要求\"帮我盯着信息流\"\"每天把新条目按来源和类型筛一遍\"\"给我一张信息流跟踪表\"时使用。按来源与冲击类型筛选条目流，输出条目清单、来源计数、期间对比与来源健康度水位；滞后与不可用如实标注\"信息可能不完整\"。不做预告、不预测影响、不做窗口统计；定时触发属宿主 cron，本技能只提供可重复执行的输入定义与输出形态。" } ], "mcpServers": [ { "registryName": "fd-info" }, { "registryName": "fd-quant" }, { "registryName": "fd-daas-mcp" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "DNYnRrjjAWINzqR_ArmbjQ", "slug": "pack-dnynrrjjawinzqr-armbjq", "category": "facade", "marketCategory": "wire", "tags": [ "柏讯", "预期差", "互动问答", "wire" ], "canonical": { "brief": "Wire Expectation Gap four-skill set: interactive Q&A extraction (structures info-stream sse-interaction items into 'investor question + company answer' pairs, each with item_id, verbatim quotes and body-status triage), expectation-clue reading (five clue classes — guidance / confirmation / denial / " }, "human": { "zh-CN": { "name": "柏讯 · 预期差", "summary": "柏讯预期差四件套：互动问答提取（info-stream 的 sse-interaction 条目结构化为「投资者问题 + 公司答复」，逐条带 item_id、逐字片段与正文状态分流）、预期线索解读（指引/确认/否认/回避/措辞变化五类线索，每条带可回原文的证据，措辞变化须双条并排，线索是公司口径不是事实）、与业绩预告对照（T_qa vs T_pre vs T_act 三段时间轴逐报告期对照，数值算术只走 fd-quant 计算面，预告区间三口径并列）、缺口陈述（已验证/未验证/不可评三档合成，每条带证据链与样本量，缺样本标注「不作依据」）。覆盖边界（上游实调 2026-10-23）：上证 e 互动可用、深交所互动易上游不可用（旧 API 405、新版面 SPA 化），本套件覆盖=SSE 上市公司，深市公司不在能力内，每次输出必须声明该缺口。核心纪律：引用任何问答前先查来源健康度；禁止模型记忆补问答；结论带样本量与时间窗；不评分、不排序、不推荐、不预测。与 info-shock 套件的分工：那边做信息条目的价格影响窗口，本套件只做公司口径与自身披露的对照，不涉及价格。不构成投资建议。" }, "en": { "name": "Wire · Expectation Gap", "summary": "Wire Expectation Gap four-skill set: interactive Q&A extraction (structures info-stream sse-interaction items into 'investor question + company answer' pairs, each with item_id, verbatim quotes and body-status triage), expectation-clue reading (five clue classes — guidance / confirmation / denial / evasion / wording change — each backed by evidence traceable to the original text; wording changes always shown as a side-by-side pair; a clue is company positioning, not fact), preannouncement cross-check (a three-segment timeline per report period — Q&A date vs preannouncement/express report vs actuals; all arithmetic goes through the fd-quant compute face; preannouncement ranges shown as lower/upper/mid), and gap statement (synthesis in three tiers — verified / unverified / not assessable — each statement with an evidence chain and sample size; missing samples are marked 'not to be relied on'). Coverage boundary (upstream live probe 2026-10-23): SSE e-interaction is available; the SZSE interaction platform is unreachable upstream (legacy API 405, new portal SPA-walled), so coverage = SSE-listed companies only, SZSE names are out of scope, and every output must state this gap. Core discipline: check source health before citing any Q&A; never fill Q&A from model memory; conclusions carry sample size and time window; no scoring, no ranking, no recommendations, no forecasts. Division of labor with the info-shock suite: that one handles price-impact windows of information items, this one only contrasts company positioning against the company's own disclosures and touches no prices. Not investment advice." }, "es": { "name": "Wire · Brecha de expectativas", "summary": "Suite Wire de brecha de expectativas en cuatro habilidades: extracción de preguntas y respuestas interactivas (estructura los ítems sse-interaction de info-stream en pares «pregunta del inversor + respuesta de la compañía», cada uno con item_id, citas literales y estado del cuerpo), lectura de pistas de expectativas (cinco clases de pistas —guía / confirmación / negación / evasión / cambio de formulación—, cada una con evidencia verificable en el texto original; los cambios de formulación se muestran siempre en paralelo; una pista es postura de la compañía, no un hecho), contraste con previsiones de resultados (línea temporal de tres tramos por periodo reportado: Q&A vs previsión/informe exprés vs valores reales; la aritmética pasa siempre por el plano de cálculo fd-quant; los rangos de previsión se muestran como mínimo/medio/máximo) y declaración de brecha (síntesis en tres niveles —verificado / no verificado / no evaluable—, cada declaración con cadena de evidencia y tamaño muestral; sin muestra suficiente se marca «no sirve como fundamento»). Límite de cobertura (prueba en vivo aguas arriba, 2026-10-23): la plataforma de interacción de la SSE está disponible; la de la SZSE no está accesible aguas arriba (API antigua 405, nuevo portal SPA): la cobertura de esta suite son las cotizadas de la SSE; las compañías de Shenzhen quedan fuera, y cada salida debe declarar este límite. Disciplina central: comprobar la salud de la fuente antes de citar cualquier Q&A; no rellenar Q&A desde la memoria del modelo; cada conclusión lleva tamaño muestral y ventana temporal; sin puntuaciones, sin clasificaciones, sin recomendaciones, sin pronósticos. Reparto con la suite info-shock: aquella cubre la ventana de impacto en precios de los ítems informativos; esta solo contrasta la postura de la compañía con sus propias divulgaciones y no toca precios. No es asesoramiento de inversión." }, "fr": { "name": "Wire · Écart d'attentes", "summary": "Suite Wire « écart d'attentes » en quatre compétences : extraction des questions-réponses interactives (structure les entrées sse-interaction d'info-stream en paires « question de l'investisseur + réponse de la société », chacune avec item_id, citations verbatim et statut du corps), lecture des indices d'attentes (cinq familles d'indices — orientations / confirmations / dénégations / esquives / changements de formulation —, chacun adossé à une preuve retrouvable dans le texte original ; les changements de formulation s'affichent toujours en vis-à-vis ; un indice est une position de la société, pas un fait), confrontation aux pré-annonces de résultats (chronologie en trois segments par exercice : Q&R vs pré-annonce/communiqué rapide vs valeurs réelles ; l'arithmétique passe exclusivement par le plan de calcul fd-quant ; les fourchettes de pré-annonce figurent en basse/haute/médiane) et constat d'écart (synthèse en trois niveaux — vérifié / non vérifié / non évaluable —, chaque constat portant une chaîne de preuves et une taille d'échantillon ; sans échantillon suffisant, mention « ne sert pas de fondement »). Périmètre (test amont du 2026-10-23) : la plateforme d'interaction de la SSE est disponible ; celle de la SZSE est inaccessible en amont (ancienne API 405, nouveau portail SPA) — la suite ne couvre que les sociétés cotées à la SSE ; les valeurs de Shenzhen sont hors périmètre et chaque sortie doit le déclarer. Discipline centrale : vérifier la santé de la source avant de citer une Q&R ; ne jamais combler une Q&R depuis la mémoire du modèle ; chaque conclusion porte taille d'échantillon et fenêtre temporelle ; ni notation, ni classement, ni recommandation, ni prévision. Répartition avec la suite info-shock : celle-ci traite la fenêtre d'impact prix des items d'information, celle-là confronte uniquement la position de la société à ses propres divulgations, sans toucher aux prix. Ne constitue pas un conseil en investissement." }, "ja": { "name": "Wire · 期待差", "summary": "Wire 期待差 4 技能セット：Q&A抽出（info-stream の sse-interaction 項目を「投資家の質問＋会社の回答」ペアに構造化し、item_id・逐語引用・本文状態を 1 件ごとに付す）、期待手がかりの読解（ガイダンス／確認／否定／回避／表現変化の 5 系統の手がかりを抽出し、各手がかりに原文へ遡れる根拠を必ず付す。表現変化は 2 件を並べて提示。手がかりは会社の見解であって事実ではない）、業績予告との突合（Q&A 公開日 vs 予告・快報 vs 実績値の 3 区間タイムラインを報告期ごとに対照し、数値算術は fd-quant 計算面のみを経由、予告レンジは下限・上限・中値を併記）、ギャップ・ステートメント（検証済み／未検証／評価不能の 3 段階で合成し、各記述に根拠チェーンとサンプル数を添え、サンプル不足は「判断材料としない」と明記）。カバー範囲（2026-10-23 の上流実測）：上証 e インタラクションは利用可、深交所のインタラクション基盤は上流で利用不可（旧 API 405、新ポータルは SPA 化）——本スイートの対象は SSE 上場企業のみで、深市企業は範囲外。毎回の出力でこの限界を明示する。中核規律：Q&A を引用する前に必ずソース健全性を確認、モデルの記憶で Q&A を補わない、結論にはサンプル数と時間窓を付す、採点・順位付け・推奨・予測をしない。info-shock スイートとの分担：あちらは情報項目の価格影響窓を扱い、こちらは会社の見解と自社開示の対照のみを行い、価格には触れない。投資助言ではない。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "gap-qa-extraction", "description": "柏讯预期差 · 互动问答提取。当用户要求\"这家公司在互动平台上被问了什么、怎么答的\"\"把某段时间的投资者问答整理出来\"\"公司在 e 互动上说过什么\"时使用。从 info-stream 拉 source=sse-interaction 的条目并结构化为\"投资者问题 + 公司答复\"问答对，逐条附 item_id、发布时间、逐字片段与正文状态。覆盖边界必须随输出声明：仅上交所（上证 e 互动）上市公司；深交所互动易上游不可用，深市公司问答不在覆盖内。只做提取：不判断答复口径强弱（交 gap-signal-reading）、不对照业绩预告（交 gap-forecast-check）。" }, { "name": "gap-signal-reading", "description": "柏讯预期差 · 预期线索解读。当用户要求\"公司答复里有没有业绩指引\"\"口径跟以前比变没变\"\"哪些问题公司在回避\"\"公司在互动平台上确认过什么\"时使用。从互动问答的公司答复中抽取五类预期线索：指引类（业绩/订单/产能/时间表）、确认、否认、回避（问而不答或答非所问）、措辞变化（同一主题跨期答复对照）。每条线索必须附逐字片段与 item_id；措辞变化必须两条并排。只描述口径不判断好坏：不给\"预期差评分\"、不对照业绩预告（交 gap-forecast-check）、不合成结论（交 gap-statement）。" }, { "name": "gap-forecast-check", "description": "柏讯预期差 · 与业绩预告对照。当用户要求\"公司在互动平台的说法和后来的业绩预告对得上吗\"\"问答是在预告前还是预告后\"\"公司答复口径与实际业绩差多少\"时使用。把问答时间点放到披露时间轴上：问答发布日 vs 业绩预告/快报披露日 vs 实际值披露日，三段时间轴逐报告期对照；数值对照（口径偏离、增速差）一律传值给 fd-quant 计算面，不得心算。每张对照表必附样本量与时间窗；问答覆盖仅 SSE，深市公司不参与对照。" }, { "name": "gap-statement", "description": "柏讯预期差 · 缺口陈述。当用户要求\"这家公司的预期差结论\"\"问答口径和业绩对得上一句话怎么说\"\"给一份预期差摘要\"时使用。把前三技能的产出（问答清单、线索表、对照表）合成为限定性预期差陈述：每条陈述必须同时带证据链（问答 item_id + 逐字片段 + 披露锚点）、样本量与时间窗；已验证（有披露对照支撑）与未验证（仅有口径线索）严格分列；缺样本标注「不作依据」。不评分、不排序、不推荐、不预测。" } ], "mcpServers": [ { "registryName": "fd-info" }, { "registryName": "fd-cn-report" }, { "registryName": "fd-quant" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "1mAJYooWE-SajhzqTNMhBw", "slug": "pack-1majyoowe-sajhzqtnmhbw", "category": "facade", "marketCategory": "text", "tags": [ "文本引擎", "剧本", "古典戏曲", "公版经典", "text" ], "canonical": { "brief": "Retrieval and research over public-domain classic plays: structure analysis (act/scene skeleton and plot progression), character search (line aggregation and character-portrait analysis), scene search (locate by description and close-read a whole scene), and similar passages (semantic-vector nearest" }, "human": { "zh-CN": { "name": "文本引擎 · 剧本经典套件", "summary": "公版经典剧本的检索与研究：结构分析（幕场/折出骨架与情节推进）、角色检索（台词聚合与形象分析）、场景检索（按描述定位并细读整场）、相似段落（语义向量找近邻，跨用词差异找情境）。语料覆盖莎士比亚、希腊悲剧、元杂剧、明清传奇、现代话剧（老舍）、京剧，共 18 部；每个结果携带 provenance 三件套（source/stable_id/as_of）。检索走 fd-text-core 引擎的混合通道（关键词 CJK bigram + 繁简归一 ⊕ 语义向量 Qwen3-Embedding-8B）。" }, "en": { "name": "Text Engine · Screenplay Classics Suite", "summary": "Retrieval and research over public-domain classic plays: structure analysis (act/scene skeleton and plot progression), character search (line aggregation and character-portrait analysis), scene search (locate by description and close-read a whole scene), and similar passages (semantic-vector nearest neighbors that find matching situations across differing wording). The corpus covers 18 works — Shakespeare, Greek tragedy, Yuan zaju, Ming–Qing chuanqi, modern spoken drama (Lao She), and Peking opera; every result carries the provenance triplet (source/stable_id/as_of). Retrieval runs through the fd-text-core engine's hybrid channel (keywords CJK bigram + traditional/simplified normalization ⊕ semantic vectors via Qwen3-Embedding-8B)." }, "es": { "name": "Text Engine · Suite de clásicos dramáticos", "summary": "Búsqueda e investigación sobre obras clásicas de dominio público: análisis estructural (esqueleto de actos/escenas y progresión del argumento), búsqueda de personajes (agregación de réplicas y análisis del retrato del personaje), búsqueda de escenas (localizar por descripción y leer en detalle la escena completa) y pasajes similares (vecinos por vectores semánticos, que encuentran situaciones afines más allá de las diferencias de vocabulario). El corpus cubre 18 obras: Shakespeare, tragedia griega, zaju del período Yuan, chuanqi de Ming–Qing, teatro hablado moderno (Lao She) y ópera de Pekín; cada resultado lleva el trío de procedencia (source/stable_id/as_of). La búsqueda usa el canal híbrido del motor fd-text-core (palabras clave CJK bigram + normalización tradicional/simplificada ⊕ vectores semánticos con Qwen3-Embedding-8B)." }, "fr": { "name": "Text Engine · Suite de classiques dramatiques", "summary": "Recherche et étude de pièces classiques du domaine public : analyse structurelle (ossature d'actes/scènes et progression de l'intrigue), recherche de personnages (agrégation des répliques et analyse du portrait du personnage), recherche de scènes (localisation par description et lecture fine d'une scène entière) et passages similaires (plus proches voisins par vecteurs sémantiques, pour retrouver des situations au-delà des différences de vocabulaire). Le corpus couvre 18 œuvres : Shakespeare, tragédie grecque, zaju des Yuan, chuanqi des Ming–Qing, théâtre parlé moderne (Lao She) et opéra de Pékin ; chaque résultat porte le triplet de provenance (source/stable_id/as_of). La recherche passe par le canal hybride du moteur fd-text-core (mots-clés CJK bigram + normalisation traditionnel/simplifié ⊕ vecteurs sémantiques Qwen3-Embedding-8B)." }, "ja": { "name": "Text Engine · 戯曲クラシックスイート", "summary": "パブリックドメインの古典戯曲の検索と研究：構造分析（幕・場の骨格とプロットの進行）、キャラクター検索（台詞の集約と人物像の分析）、場面検索（描写から特定して一場面を丸ごと精読）、類似箇所（意味ベクトルの近傍検索で、言葉遣いが違っても同じ状況を見つける）。コーパスはシェイクスピア、ギリシア悲劇、元雑劇、明清伝奇、現代話劇（老舎）、京劇の計18作品。各結果は provenance 三点セット（source/stable_id/as_of）を携えます。検索は fd-text-core エンジンのハイブリッド経路（キーワード CJK bigram＋繁体字/簡体字正規化 ⊕ 意味ベクトル Qwen3-Embedding-8B）を使います。" } }, "whenToUse": "Browse under \"text\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "screenplay-structure-analysis", "description": "剧本结构分析（文本引擎·剧本包入口技能）。当用户想了解一部戏剧/戏曲的幕场骨架、情节推进、折出结构，或要求梳理某剧的场次脉络、剧情大纲时使用。驱动从结构骨架到情节脉络的分析流程，数据经 fd-text-core 引擎的剧本语料实时检索。" }, { "name": "screenplay-character-search", "description": "剧本角色检索与分析（文本引擎·剧本包）。当用户想查某个人物在某部剧里说了什么、出场段落、角色形象分析，或要求\"找出哈姆雷特的独白\"\"窦娥的唱词\"\"这个角色在全剧中的台词\"时使用。按角色名驱动台词级检索，数据经 fd-text-core 引擎实时取。" }, { "name": "screenplay-scene-search", "description": "剧本场景检索与细读（文本引擎·剧本包）。当用户想查某部剧的某个具体场景（\"哈姆雷特的戏中戏\"\"窦娥临刑那场\"\"西厢记的送别\"）、某个情境段落、或要求细读某场戏的完整内容与调度时使用。按场景描述驱动检索，数据经 fd-text-core 引擎实时取。" }, { "name": "screenplay-similar-passages", "description": "剧本相似段落检索与比较（文本引擎·剧本包）。当用户给出一段剧本文字想找\"类似的其他段落\"、要求跨剧对比相似情境（如\"还有哪些戏写了临刑发誓\"\"找和这段相似的独白\"）、或做母题/桥段溯源时使用。按参考文本的语义向量找近邻，数据经 fd-text-core 引擎实时取。" } ], "mcpServers": [ { "registryName": "fd-text-core" } ] } }, { "id": "1OQ93FLBzzVGcn2js31ADA", "slug": "pack-1oq93flbzzvgcn2js31ada", "category": "facade", "marketCategory": "text", "tags": [ "文本引擎", "论文", "学术检索", "arXiv", "text" ], "canonical": { "brief": "Retrieval and research over academic papers: literature review (framing the direction and mapping the thread), method tracking (lineage of proposals/variants/applications), and related work (semantically similar papers and comparison matrices). The corpus covers thousand-scale samples across eight a" }, "human": { "zh-CN": { "name": "文本引擎 · 论文研究套件", "summary": "学术论文的检索与研究：文献综述（方向定位与脉络梳理）、方法追踪（提出/变体/应用谱系）、相关工作（语义相似文献与对比矩阵）。语料覆盖 arXiv q-fin 与 cs.CL 等八个类目的千级样本（元数据 + 摘要，CC0 元数据）；外部开放检索（arXiv/OpenAlex/Semantic Scholar）编排补充。**版权红线**：只返回元数据、摘要、切块引用与指回 arXiv.org 的原文链接，本服务不提供也不外发论文全文。检索走 fd-text-core 引擎的混合通道（关键词 ⊕ 语义向量 Qwen3-Embedding-8B）。" }, "en": { "name": "Text Engine · Paper Research Suite", "summary": "Retrieval and research over academic papers: literature review (framing the direction and mapping the thread), method tracking (lineage of proposals/variants/applications), and related work (semantically similar papers and comparison matrices). The corpus covers thousand-scale samples across eight arXiv categories including q-fin and cs.CL (metadata + abstracts, CC0 metadata); external open search (arXiv/OpenAlex/Semantic Scholar) is orchestrated as a supplement. **Copyright red line**: only metadata, abstracts, chunked quotes, and links pointing back to arXiv.org are returned — this service neither provides nor sends out full paper texts. Retrieval runs through the fd-text-core engine's hybrid channel (keywords ⊕ semantic vectors via Qwen3-Embedding-8B)." }, "es": { "name": "Text Engine · Suite de investigación de artículos científicos", "summary": "Búsqueda e investigación sobre artículos académicos: revisión de literatura (encuadre de la dirección y reconstrucción del hilo), seguimiento de métodos (linaje de propuestas/variantes/aplicaciones) y trabajos relacionados (artículos semánticamente afines y matrices comparativas). El corpus cubre muestras de escala de miles en ocho categorías de arXiv, incluidas q-fin y cs.CL (metadatos + resúmenes, metadatos CC0); la búsqueda abierta externa (arXiv/OpenAlex/Semantic Scholar) se orquesta como complemento. **Línea roja de copyright**: solo se devuelven metadatos, resúmenes, citas por fragmentos y enlaces que remiten a arXiv.org; este servicio ni provee ni hace circular textos completos de los artículos. La búsqueda usa el canal híbrido del motor fd-text-core (palabras clave ⊕ vectores semánticos con Qwen3-Embedding-8B)." }, "fr": { "name": "Text Engine · Suite de recherche d'articles scientifiques", "summary": "Recherche et étude d'articles académiques : revue de littérature (cadrage de la direction et reconstitution du fil), suivi des méthodes (généalogie des propositions/variantes/applications) et travaux connexes (articles sémantiquement proches et matrices comparatives). Le corpus couvre des échantillons de l'ordre du millier dans huit catégories arXiv dont q-fin et cs.CL (métadonnées + résumés, métadonnées CC0) ; la recherche ouverte externe (arXiv/OpenAlex/Semantic Scholar) est orchestrée en complément. **Ligne rouge du droit d'auteur** : seuls sont renvoyés métadonnées, résumés, citations par extraits et liens renvoyant vers arXiv.org — ce service ne fournit ni ne diffuse les textes intégraux des articles. La recherche passe par le canal hybride du moteur fd-text-core (mots-clés ⊕ vecteurs sémantiques Qwen3-Embedding-8B)." }, "ja": { "name": "Text Engine · 論文研究スイート", "summary": "学術論文の検索とリサーチ：文献レビュー（方向づけと系譜の整理）、メソッド追跡（提案/変種/応用の系譜）、関連研究（意味的に近い文献と比較マトリクス）。コーパスは arXiv の q-fin や cs.CL など8カテゴリの千件規模サンプルをカバー（メタデータ＋要旨、CC0 メタデータ）。外部オープン検索（arXiv/OpenAlex/Semantic Scholar）をオーケストレーションして補完します。**著作権の赤線**：返すのはメタデータ・要旨・チャンク引用・arXiv.org への原文リンクのみで、本サービスは論文の全文を提供せず外部に送出もしません。検索は fd-text-core エンジンのハイブリッド経路（キーワード ⊕ 意味ベクトル Qwen3-Embedding-8B）を使います。" } }, "whenToUse": "Browse under \"text\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "paper-literature-search", "description": "论文文献检索（文本引擎·论文包入口技能）。当用户想查某个研究方向的论文、做文献综述、了解某领域有哪些工作、或要求\"找找关于 XX 的论文\"\"这个方向的研究现状\"时使用。驱动从方向定位到文献梳理的流程，数据经 fd-text-core 引擎的论文语料与外部开放检索（arXiv/OpenAlex/S2）实时获取。" }, { "name": "paper-method-tracking", "description": "论文方法追踪（文本引擎·论文包）。当用户想查某个方法/技术（如 transformer、对比学习、GARCH）在学术论文里怎么用、有哪些变体、谁提出的，或要求\"追一下 XX 方法的发展\"时使用。按方法名驱动检索与脉络梳理，数据经 fd-text-core 引擎实时取。" }, { "name": "paper-related-work", "description": "论文相关工作与相似文献（文本引擎·论文包）。当用户给出一段摘要/题目/研究想法，要求找相似论文、写相关工作段落、或做查重式比对（\"有没有做过类似的\"）时使用。按参考文本的语义向量找近邻，数据经 fd-text-core 引擎实时取。" } ], "mcpServers": [ { "registryName": "fd-text-core" } ] } }, { "id": "UXWcHRYPeyNue1Z4qQl-Tw", "slug": "pack-uxwchrypeynue1z4qql-tw", "category": "facade", "marketCategory": "business", "tags": [ "零售", "个体户", "同业密度", "批发价", "信号" ], "canonical": { "brief": "Retail industry environmental signal pack (install alongside the Micro-Business · Core Checkup Suite): wholesale price reference (Kunming International Flower Auction Center, kifa.net, daily auction prices and market analysis — a direct purchase-cost reference for florists and fresh-produce shops, n" }, "human": { "zh-CN": { "name": "生意体检 · 零售信号包", "summary": "零售行业环境信号包（需与「生意体检 · 内核套件」叠加安装）：批发价参考（昆明国际花卉拍卖中心 kifa.net 日度拍卖价与行情分析，对花店/生鲜类是进货成本直接参考，不泛化为所有零售商品）、城市基本面（中国城市面板 209 变量 × 297 城 × 2000–2024，判断社区消费力与选址）、同业密度（工商登记，支持省份+城市+区县与行业小类颗粒度，可用 company_type 筛选个体工商户）、快递大盘景气（国家邮政局快递月报，36 期连续，兼营电商/社区团购的小店参考）。**硬边界**：工商数据最新批次 2023，滞后约 3 年，一律标注「历史格局、非实时」；数据源只有新增注册口径、无注销/吊销，「关了多少家」类问题明示做不了并给存量状态分布替代，禁止编造关店数量或估算退出率。" }, "en": { "name": "Micro-Business · Retail Signal Pack", "summary": "Retail industry environmental signal pack (install alongside the Micro-Business · Core Checkup Suite): wholesale price reference (Kunming International Flower Auction Center, kifa.net, daily auction prices and market analysis — a direct purchase-cost reference for florists and fresh-produce shops, not generalizable to all retail goods), city fundamentals (China city panel, 209 variables × 297 cities × 2000–2024, for judging neighborhood purchasing power and site selection), peer density (business registry with province + city + county granularity down to sub-industry, and company_type filtering for individually-owned businesses), and express-delivery climate (State Post Bureau monthly reports, 36 consecutive issues, a reference for small shops that also sell online or via community group buying). **Hard boundaries**: the newest registry batch is 2023 — roughly a 3-year lag — so every output is labeled 'historical snapshot, not real-time'; the source covers only new registrations with no deregistration/revocation scope, 'how many shops closed' questions are explicitly declined with a stock status distribution offered instead, and fabricating closure counts or estimating exit rates is forbidden." }, "es": { "name": "Micro-Business · Pack de señales de comercio minorista", "summary": "Paquete de señales ambientales del sector minorista (instalar junto al Micro-Business · Suite de chequeo central): referencia de precios al por mayor (Centro Internacional de Subastas de Flores de Kunming, kifa.net, precios diarios de subasta y análisis de mercado — referencia directa de coste de compra para floristerías y tiendas de productos frescos, no generalizable a todos los productos minoristas), fundamentales de ciudad (panel de ciudades de China, 209 variables × 297 ciudades × 2000–2024, para juzgar el poder de compra del barrio y la elección de ubicación), densidad de competencia (registro mercantil con granularidad de provincia + ciudad + distrito hasta subcategoría sectorial, con filtro company_type para negocios unipersonales) y clima del mercado de paquetería (informes mensuales de la Oficina Estatal de Correos, 36 números consecutivos, referencia para tiendas pequeñas que también venden online o por compra grupal comunitaria). **Límites duros**: el lote más reciente del registro es 2023, con unos 3 años de retraso, así que todo se etiqueta «panorama histórico, no en tiempo real»; la fuente solo cubre altas de registro sin cómputo de bajas/revocaciones, las preguntas del tipo «cuántas tiendas cerraron» se declinan explícitamente ofreciendo en su lugar la distribución de estados del inventario, y está prohibido inventar cifras de cierres o estimar tasas de salida." }, "fr": { "name": "Micro-Business · Pack de signaux pour le commerce de détail", "summary": "Pack de signaux environnementaux pour le commerce de détail (à installer en complément du Micro-Business · Suite de diagnostic de base) : référence des prix de gros (Kunming International Flower Auction Center, kifa.net, prix d'adjudication quotidiens et analyse de marché — une référence directe du coût d'achat pour fleuristes et primeurs, non généralisable à tous les produits de détail), fondamentaux de ville (panneau des villes de Chine, 209 variables × 297 villes × 2000–2024, pour juger du pouvoir d'achat du voisinage et du choix d'emplacement), densité concurrentielle (registre du commerce à la granularité province + ville + district jusqu'à la sous-branche, avec filtre company_type pour les commerçants individuels) et climat du marché express (rapports mensuels de l'Administration postale d'État, 36 numéros consécutifs, une référence pour les petites boutiques qui vendent aussi en ligne ou en achat groupé de quartier). **Limites dures** : le lot le plus récent du registre date de 2023, avec environ 3 ans de décalage — tout est donc annoté « panorama historique, pas en temps réel » ; la source ne couvre que les immatriculations, sans périmètre de radiations/retraits, les questions du type « combien de boutiques ont fermé » sont explicitement déclinées au profit de la distribution des statuts du stock, et il est interdit d'inventer des chiffres de fermetures ou d'estimer des taux de sortie." }, "ja": { "name": "Micro-Business · 小売シグナルパック", "summary": "小売業の環境シグナルパック（「Micro-Business · コア診断スイート」との併設が必要）：卸価格リファレンス（昆明国際花卉オークションセンター kifa.net の日次落札価格と市況分析。花屋/生鮮店にとっての仕入コストの直接的な参照であり、すべての小売商品への一般化は不可）、都市ファンダメンタルズ（中国都市パネル 209変数×297都市×2000–2024。近隣の購買力と出店判断の材料）、同業密度（商業登記。省＋市＋区県と業界小類の粒度、company_type で個人事業主を絞り込み可能）、宅配便の市況（国家郵政局の月次速報、36号連続。EC/コミュニティグループ購入を兼業する小店の参考）。**ハードな境界**：登記データの最新バッチは2023年で約3年の遅延があり、一律「履歴スナップショット、非リアルタイム」と表示。データソースは新規登録の集計のみで註銷/取消の範囲がなく、「何店閉まったか」系の質問には明示的に対応不能と伝え、在庫の状態分布を代替として示します。閉店数の捏造や退出率の推定は禁止です。" } }, "whenToUse": "Browse under \"business\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "mb-signal-retail", "description": "零售行业环境信号包（谦面·生意体检姊妹件，需与内核套件 mb-core 叠加安装）。读四类信号：批发价参考（昆明花拍日度价，花店/生鲜进货直接参考）、城市基本面（中国城市面板 209 变量×297 城，判断社区消费力与选址）、同业密度（工商登记区县+行业小类，可筛个体工商户）、快递大盘景气（邮政局月报，兼营电商参考）。工商数据最新批次 2023、滞后约 3 年，一律标注「历史格局、非实时」；无注销口径，「关了多少家」明示做不了并给存量状态分布替代，禁止编造关店量。" } ], "mcpServers": [ { "registryName": "fd-find-data-business-mcp" }, { "registryName": "fd-open-data-mcp" } ] } }, { "id": "BniA5thfDdb3oG-FT0CkJA", "slug": "pack-bnia5thfddb3og-ft0ckja", "category": "facade", "marketCategory": "business", "tags": [ "餐饮", "个体户", "食材成本", "商圈", "信号" ], "canonical": { "brief": "Catering industry environmental signal pack (install alongside the Micro-Business · Core Checkup Suite): food-cost trend (Dalian Commodity Exchange futures — live hogs/soybean meal/corn/palm oil — trend warnings for catering's largest variable-cost item, not market spot prices), a commercial-distric" }, "human": { "zh-CN": { "name": "生意体检 · 餐饮信号包", "summary": "餐饮行业环境信号包（需与「生意体检 · 内核套件」叠加安装）：食材成本趋势（大连商品交易所期货：生猪/豆粕/玉米/棕榈油，餐饮最大变动成本项的趋势预警，不等于菜市场现货价）、商圈活跃度代理（专资办电影票房日榜/周榜，2017 起可回溯，反映商圈消费活跃度而非本店客流）、堂食外卖切换代理（中央气象台天气，34 省 2529 站时级，结构性提示而非客流预测），并可选查同业密度（工商登记区县 + 行业小类）。每个信号均标注代理性质与局限；信号源不可用时标注「未连接」且不填模型记忆值。工商类信号带批次年份与「历史格局、非实时」标注，且明示无注销/吊销口径、不做关店量。" }, "en": { "name": "Micro-Business · Catering Signal Pack", "summary": "Catering industry environmental signal pack (install alongside the Micro-Business · Core Checkup Suite): food-cost trend (Dalian Commodity Exchange futures — live hogs/soybean meal/corn/palm oil — trend warnings for catering's largest variable-cost item, not market spot prices), a commercial-district activity proxy (national film box office daily/weekly charts, traceable back to 2017; reflects district consumption activity, not your store's footfall), and a dine-in/delivery shift proxy (Central Meteorological Observatory weather, hourly across 34 provinces and 2,529 stations; a structural hint, not a footfall forecast), plus optional peer-density lookup (business registry at county + sub-industry granularity). Every signal is labeled with its proxy nature and its limits; when a source is unavailable it is marked 'not connected' and never filled with model-memory values. Registry-based signals carry their batch year with a 'historical snapshot, not real-time' label, and explicitly state that there is no deregistration/revocation scope and no closure counts." }, "es": { "name": "Micro-Business · Pack de señales de restauración", "summary": "Paquete de señales ambientales del sector restauración (instalar junto al Micro-Business · Suite de chequeo central): tendencia de costes de alimentos (futuros de la Bolsa de Mercancías de Dalian —cerdo vivo/harina de soja/maíz/aceite de palma—, alertas de tendencia sobre la mayor partida de coste variable de la restauración; no son precios spot de mercado), proxy de actividad de la zona comercial (listas diarias/semanales de taquilla cinematográfica, rastreables desde 2017; refleja la actividad de consumo de la zona, no la clientela de tu tienda) y proxy de cambio entre servicio en local y a domicilio (tiempo del Observatorio Meteorológico Central, horario, 34 provincias y 2529 estaciones; una indicación estructural, no una previsión de clientela), más consulta opcional de densidad de competencia (registro mercantil con granularidad de distrito + subcategoría sectorial). Cada señal indica su carácter de proxy y sus límites; si la fuente no está disponible se marca «no conectado» y nunca se rellena con valores de memoria del modelo. Las señales del registro llevan año de lote con la marca «panorama histórico, no en tiempo real», y declaran explícitamente que no hay cómputo de bajas/revocaciones ni cifras de cierres." }, "fr": { "name": "Micro-Business · Pack de signaux pour la restauration", "summary": "Pack de signaux environnementaux pour la restauration (à installer en complément du Micro-Business · Suite de diagnostic de base) : tendance des coûts des denrées (futurs de la Bourse des marchandises de Dalian — porcs vivants/tourteau de soja/maïs/huile de palme — alertes de tendance sur le plus gros poste de coûts variables de la restauration, pas des prix au comptant de marché), proxy d'activité de la zone commerciale (palmarès quotidiens/hebdomadaires du box-office cinématographique, rétroactifs depuis 2017 ; reflète l'activité de consommation du quartier, pas le passage dans votre boutique) et proxy de bascule entre sur place et livraison (météo de l'Observatoire météorologique central, à l'heure, 34 provinces et 2 529 stations ; une indication structurelle, pas une prévision de fréquentation), plus consultation optionnelle de la densité concurrentielle (registre du commerce à la granularité district + sous-branche). Chaque signal est annoté de sa nature de proxy et de ses limites ; si la source est indisponible il est marqué « non connecté » et jamais rempli de valeurs sorties de la mémoire du modèle. Les signaux issus du registre portent l'année de lot avec la mention « panorama historique, pas en temps réel », et indiquent explicitement l'absence de périmètre de radiations/retraits et l'absence de chiffres de fermetures." }, "ja": { "name": "Micro-Business · 飲食シグナルパック", "summary": "飲食業の環境シグナルパック（「Micro-Business · コア診断スイート」との併設が必要）：食材コスト傾向（大連商品交易所先物：生豚/大豆ミール/トウモロコシ/パーム油。飲食業最大の変動費項目のトレンド警報で、市場のスポット価格ではない）、商圏の賑わいプロキシ（映画興行収入の日次/週次ランキング、2017年以降に遡及可能。商圏の消費活性を反映し、自店の来客数ではない）、イートイン/デリバリー切替プロキシ（中央気象台の天気、34省2529地点の時間粒度。構造的なヒントであって来客予測ではない）、加えて同業密度の照会（任意：商業登記の区県＋業界小類）。各シグナルにはプロキシ性質と限界を必ず注記。シグナルソースが使えない場合は「未接続」と表示し、モデル記憶の値で埋めません。登記系シグナルにはバッチ年と「履歴スナップショット、非リアルタイム」を付け、註銷/取消の集計範囲がなく閉店数は出せないことを明示します。" } }, "whenToUse": "Browse under \"business\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "mb-signal-catering", "description": "餐饮行业环境信号包（谦面·生意体检姊妹件，需与内核套件 mb-core 叠加安装）。读三类信号：食材成本趋势（大商所期货生猪/豆粕/玉米/棕榈油，餐饮最大变动成本项）、商圈活跃度代理（专资办电影票房日榜/周榜）、堂食外卖切换代理（中央气象台天气），并可选查同业密度（工商登记区县+行业小类）。每个信号标注代理性质与局限，源不可用即如实标注未连接，不填模型记忆值。工商类带批次年份与「历史格局、非实时」，且明示无注销口径、不做关店量。" } ], "mcpServers": [ { "registryName": "fd-find-data-business-mcp" }, { "registryName": "fd-open-data-mcp" } ] } }, { "id": "MBcXTHtyyMOxDHmSUYvDdQ", "slug": "pack-mbcxthtyymoxdhmsuyvddq", "category": "facade", "marketCategory": "business", "tags": [ "个体户", "微型企业", "生意体检", "盈亏平衡", "经营决策" ], "canonical": { "brief": "Operating-checkup core for China's micro-enterprises and sole proprietors: a four-stage workflow (intake interview → single-store accounting → industry-signal comparison → action list) as the entry skill, with three companion calculators for break-even, gross-margin structure, and cash flow/payment " }, "human": { "zh-CN": { "name": "生意体检 · 内核套件", "summary": "面向中国传统微型企业与个体户的经营体检内核：四段式工作流（问诊建档 → 单店算账 → 行业信号对照 → 行动清单）为入口技能，配套盈亏平衡、毛利结构、现金流账期三个计算器。所有金额与比率一律经计算器产出并附计算公式 + 输入基数 + 来源（来源名/地区/年份/口径时间），外部统计基数取不到时该项中止并明示「基数数据源未连接/未定码」，禁止用模型记忆的行业数字填充。未安装行业信号包时如实标注「未安装行业信号包」且不中断分析。单店生意档案可建立在 DAAS 实体模型上（一家店=一个实体，指标持久化 + 告警规则）。内核按小老板能听懂的标准书写，专业术语附日常解释；机构与政务渠道共用同一内核，仅外壳与推送方式不同。" }, "en": { "name": "Micro-Business · Core Checkup Suite", "summary": "Operating-checkup core for China's micro-enterprises and sole proprietors: a four-stage workflow (intake interview → single-store accounting → industry-signal comparison → action list) as the entry skill, with three companion calculators for break-even, gross-margin structure, and cash flow/payment terms. Every amount and ratio is produced by a calculator and carries its formula, input base, and source (source name/region/year/as-of date); when an external statistical base cannot be fetched, that item halts with an explicit 'base data source not connected/uncoded' notice, and filling gaps from the model's memory of industry figures is forbidden. When no industry signal pack is installed, this is honestly labeled and the analysis continues. Single-store business profiles can be persisted on the DAAS entity model (one store = one entity, with persistent indicators and alert rules). The core is written so small-business owners can understand it, with plain-language explanations for technical terms; institutional and government channels share the same core and differ only in shell and delivery." }, "es": { "name": "Micro-Business · Suite de chequeo central", "summary": "Núcleo de chequeo operativo para microempresas y negocios unipersonales de China: un flujo de cuatro etapas (entrevista de diagnóstico → cuentas de la tienda → contraste con señales del sector → lista de acciones) como habilidad de entrada, con tres calculadoras acompañantes: punto de equilibrio, estructura de margen bruto y flujo de caja/plazos de pago. Todo importe y ratio se produce con una calculadora y lleva fórmula de cálculo, base de entrada y fuente (nombre de la fuente/región/año/corte temporal); si no se puede obtener una base estadística externa, el ítem se detiene e indica explícitamente «fuente de datos base no conectada/sin código», y está prohibido rellenar con cifras sectoriales memorizadas por el modelo. Sin paquete de señales sectoriales instalado, se declara honestamente y el análisis no se interrumpe. El perfil de cada tienda puede persistirse en el modelo de entidades DAAS (una tienda = una entidad, con indicadores persistentes y reglas de alerta). El núcleo se redacta para que el pequeño comerciante lo entienda, con explicaciones cotidianas de los términos técnicos; los canales institucionales y gubernamentales comparten el mismo núcleo y solo cambian la envoltura y el modo de entrega." }, "fr": { "name": "Micro-Business · Suite de diagnostic de base", "summary": "Noyau de diagnostic opérationnel pour les micro-entreprises et travailleurs indépendants de Chine : un flux en quatre étapes (entretien de diagnostic → comptes du magasin → confrontation aux signaux sectoriels → liste d'actions) comme compétence d'entrée, avec trois calculatrices compagnes : seuil de rentabilité, structure de marge brute et trésorerie/délais de paiement. Tout montant et tout ratio sortent d'une calculatrice et portent leur formule, leur base d'entrée et leur source (nom de la source/région/année/date de référence) ; si une base statistique externe est introuvable, l'item s'arrête en signalant explicitement « source de données de base non connectée/non codée », et il est interdit de combler avec des chiffres sectoriels sortis de la mémoire du modèle. Sans pack de signaux sectoriels installé, cela est déclaré honnêtement et l'analyse continue. La fiche de chaque magasin peut être persistée sur le modèle d'entités DAAS (un magasin = une entité, avec indicateurs persistants et règles d'alerte). Le noyau est rédigé pour être compris des petits patrons, avec des explications en langage courant pour chaque terme technique ; les canaux institutionnels et gouvernementaux partagent le même noyau, seuls l'habillage et le mode de diffusion changent." }, "ja": { "name": "Micro-Business · コア診断スイート", "summary": "中国の零細企業・個人事業主向けの経営診断コア：四段ワークフロー（ヒアリング記録 → 単店の損益計算 → 業界シグナル対照 → アクションリスト）を入口スキルとし、損益分岐点・粗利構造・キャッシュフロー/支払サイトの3つの計算機を同梱。金額と比率はすべて計算機が算出し、計算式・入力値・出典（ソース名/地域/年/基準時点）を必ず添付します。外部統計の基礎値が取得できない場合はその項目を中止し「基礎データソース未接続/未コード化」を明示、モデル記憶による業界数値の埋め合わせは禁止です。業界シグナルパック未導入時はその旨を正直に表示し、分析は中断しません。単一店舗の経営プロフィールは DAAS エンティティモデル上に永続化できます（1店=1エンティティ、指標の永続化＋アラートルール）。コアは小さなオーナーに伝わる言葉で書かれ、専門用語には平易な解説を付けます。法人・行政の各チャネルも同一コアを使い、外装と配信方法のみが異なります。" } }, "whenToUse": "Browse under \"business\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "mb-business-checkup", "description": "生意体检（谦面·面向个体户与微型企业的内核入口技能）。当小老板描述自己的店（餐馆/便利店/花店/网店）、或代账、银行普惠客户经理、园区运营要为商户做经营诊断时使用。四段式：问诊建档 → 单店算账（点名盈亏平衡/毛利结构/现金流账期三计算器）→ 行业信号对照 → 行动清单。数值只由计算器产出并附公式与来源；未装行业信号包时如实标注，不中断分析。" }, { "name": "mb-breakeven-calculator", "description": "盈亏平衡计算器（谦面·面向个体户与微型企业的三个原子计算器之一）。当小老板或代账要算「一个月卖到多少才不亏」「每天要来多少客人/多少单才保本」「离保本线还有多远」「按现在的流水多少天能 cover 固定成本」时使用。产出月盈亏平衡流水、保本客流/单量、安全边际、保本天数，每个数字附计算公式、输入基数与来源标注。只做计算，不做问诊。" }, { "name": "mb-margin-structure-calculator", "description": "毛利结构计算器（谦面·面向个体户与微型企业的三个原子计算器之一）。当小老板或代账要算「毛利有多少」「哪块成本占得最重」「房租人工占流水几成」「涨价能多赚多少」时使用。产出固定/变动成本拆分、各项成本占流水比、毛利率与边际贡献率、成本结构诊断。每个数字附计算公式、输入基数与来源标注；外部基数取不到即中止该项，禁止用模型记忆的行业数字填充。只做计算，不做问诊。" }, { "name": "mb-cashflow-terms-calculator", "description": "现金流与账期计算器（谦面·面向个体户与微型企业的三个原子计算器之一）。当小老板或代账要算「手里要留多少周转金」「什么时候会缺钱」「供应商账期和客户账期对不上怎么办」「旺季备货钱够不够」时使用。产出最少周转金、现金流缺口时段与金额、账期占款、季节性缓冲建议。每个数字附计算公式、输入基数与来源标注；外部基数取不到即中止该项，禁止用模型记忆的行业数字填充。只做计算，不做问诊。" } ], "mcpServers": [ { "registryName": "fd-find-data-business-mcp" }, { "registryName": "fd-daas-mcp" } ] } }, { "id": "3uOELnBMA6WiWSTQxrV93g", "slug": "pack-3uoelnbma6wiwstqxrv93g", "category": "facade", "marketCategory": "legal", "tags": [ "识律", "法律", "lex" ], "canonical": { "brief": "Civil/commercial litigation drafting and filing prep: complaints, defenses and evidence lists, with evidence cards and checklists; v2 adds element-based format conversion (traditional documents → 67 SPC model templates, per-element source tags, facts preserved). Statute anchoring, validity checks an" }, "human": { "zh-CN": { "name": "识律 · 诉讼文书套件", "summary": "民商事诉讼文书起草与立案材料整理：起诉状/答辩状/证据目录生成，配套证据卡片与材料清单整理；v2 新增**要素式格式转换**（传统文书→最高院 67 类示范文本，逐要素来源标注，事实保真）。法条锚定、效力校验、类案与指导案例参照全部走识律法律检索数据源，案号与条文一律来自真实检索返回。" }, "en": { "name": "Lex · Litigation Documents Suite", "summary": "Civil/commercial litigation drafting and filing prep: complaints, defenses and evidence lists, with evidence cards and checklists; v2 adds element-based format conversion (traditional documents → 67 SPC model templates, per-element source tags, facts preserved). Statute anchoring, validity checks and similar/guiding-case references run on Lex legal search; case numbers and provision text come from real search returns." }, "es": { "name": "Lex · Suite de escritos de litigación", "summary": "Escritos de litigación civil y mercantil y preparación del expediente: demandas, contestaciones y listas de pruebas, con tarjetas de prueba y listas; la v2 añade la conversión por elementos (documentos tradicionales → 67 modelos del Tribunal Supremo, fuente etiquetada por elemento, hechos fieles). Anclaje de artículos, vigencia y casos análogos/guía van por la búsqueda jurídica Lex; números de causa y textos provienen de búsquedas reales." }, "fr": { "name": "Lex · Suite d'actes de procédure", "summary": "Actes de procédure civils et commerciaux et préparation du dossier : requêtes, conclusions en défense et listes de pièces, avec fiches de preuves et listes ; la v2 ajoute la conversion par éléments (actes traditionnels → 67 modèles types de la Cour suprême, source étiquetée par élément, faits préservés). Ancrage des textes, validité et affaires analogues/guide passent par la recherche Lex ; numéros et textes viennent de recherches réelles." }, "ja": { "name": "Lex · 訴訟文書スイート", "summary": "民商事訴訟文書の起案と立件資料の整理：訴状・答弁書・証拠目録の生成、証拠カードと資料リストの整理。v2 では要素式フォーマット変換（従来文書→最高裁 67 種のモデル書式、要素ごとの出所注記、事実の忠実性）を追加。法条アンカー・効力検証・類似事案と指導案例の参照はすべて Lex の法律検索データソースから、事件番号と条文は実検索の返却値による。" } }, "whenToUse": "Browse under \"legal\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "litigation-drafting", "description": "民商事诉讼文书起草（起诉状/答辩状/证据目录）。当用户提供案情或案件材料，要求写起诉状、答辩状、上诉状要点、证据目录或立案配套文书时使用。法条锚定与类案参照均走真实检索数据源，律师确认要点后才生成成稿。" }, { "name": "evidence-card-organizer", "description": "证据卡片提取与材料清单整理（立案材料整理技能）。当用户提交一批案件材料（合同、转账记录、聊天记录、票据、送达回证等），要求提取证据、编证据目录、整理材料清单或做证据关系与矛盾排查时使用。输出双视图证据卡片库与待补强清单，供起诉状/答辩状起草直接引用。" }, { "name": "format-conversion-elementized", "description": "要素式格式转换。当用户上传或粘贴一份传统式法律文书（起诉状、答辩状等），希望转换为法院推广的要素式示范文本格式时使用。识别案由并匹配最高院示范文本，逐要素从原文提取填充，逐要素标注来源（取自原文/原文缺失/需补充），原文事实保真不改写；案由无法识别或示范文本未收录时如实说明，不以自拟模板冒充示范文本。" } ], "mcpServers": [ { "registryName": "fd-legal-search-mcp" }, { "registryName": "fd-find-data-business-mcp" } ] } }, { "id": "ndVkyAg2NfIVXg7oSDAWxw", "slug": "pack-ndvkyag2nfivxg7osdawxw", "category": "facade", "marketCategory": "legal", "tags": [ "识律", "法律", "lex" ], "canonical": { "brief": "Full labor-dispute workflow plus two calculators (Lex × Wire convergence): labor arbitration end to end (application / defense / evidence list / finality test / deadline table, with an adjudicator branch) as the entry skill, plus the labor-compensation and personal-injury calculators as atomic skill" }, "human": { "zh-CN": { "name": "识律 · 劳动法套件", "summary": "劳动争议全流程 + 双计算器（识律×柏讯双线合流）：劳动仲裁全流程（申请书/答辩书/证据目录/一裁终局判断/时限表，含裁判者分支）为入口技能，劳动补偿金计算器与人身损害赔偿计算器为两个原子技能。法律依据走识律条级引用解析与效力核验，赔偿/补偿的统计基数走柏讯分省分年统计读取并强制 provenance（计算公式 + 基数来源 + 年份），统计源不可用或概念码未定档时中止数值计算。" }, "en": { "name": "Lex · Labor Law Suite", "summary": "Full labor-dispute workflow plus two calculators (Lex × Wire convergence): labor arbitration end to end (application / defense / evidence list / finality test / deadline table, with an adjudicator branch) as the entry skill, plus the labor-compensation and personal-injury calculators as atomic skills. Legal basis via Lex article-level citation parsing and validity checks; statistical bases via Wire province-year statistics with mandatory provenance (formula + base source + year); computation halts if the statistics source is unavailable or the concept code unset." }, "es": { "name": "Lex · Suite de derecho laboral", "summary": "Flujo completo de conflictos laborales + dos calculadoras (convergencia Lex × Wire): el arbitraje laboral completo (solicitud / contestación / pruebas / laudo único / plazos, con rama de árbitro) como entrada, más las calculadoras de indemnización laboral y de daño personal como atómicas. Base jurídica vía análisis de citas por artículo de Lex y vigencia verificada; bases estadísticas vía estadísticas Wire por provincia y año con provenance obligatoria (fórmula + fuente de la base + año); el cálculo se detiene si falta la fuente estadística o el código de concepto." }, "fr": { "name": "Lex · Suite de droit du travail", "summary": "Parcours complet des conflits du travail + deux calculateurs (convergence Lex × Wire) : l'arbitrage du travail complet (requête / défense / liste des pièces / test de sentence définitive / tableau des délais, avec branche juge) en entrée, plus les calculateurs d'indemnité de rupture et de préjudice corporel en compétences atomiques. Fondement juridique via l'analyse article par article de Lex et validité vérifiée ; bases statistiques via les statistiques Wire par province et par an avec provenance obligatoire (formule + source + année) ; calcul interrompu si la source statistique manque ou le code de concept n'est pas défini." }, "ja": { "name": "Lex · 労働法スイート", "summary": "労働紛争の全工程＋2 つの計算機（Lex×Wire の二線統合）：労働仲裁の全工程（申立書／答弁書／証拠目録／一裁終局の判断／期限表、裁定者分岐を含む）を入口スキルに、労働補償金計算機と人身損害賠償計算機を原子スキルとする。法的根拠は Lex の条文単位引用解析と効力検証、賠償・補償の統計基数は Wire の省別・年別統計の読み取りで provenance（計算式＋基数の出所＋年）を必須化。統計ソースが利用不可または概念コード未定義の場合は数値計算を中止。" } }, "whenToUse": "Browse under \"legal\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "labor-arbitration-workflow", "description": "劳动争议仲裁全流程（识律劳动法套件入口技能）。当用户描述劳动争议案情（违法解除/经济补偿/加班费/未签合同二倍工资/调岗降薪/工伤待遇）、要求仲裁申请书或答辩书、证据目录、一裁终局判断、仲裁时限表，或以裁判者身份要求争议焦点归纳与裁决说理时使用。按立场（申请人/被申请人/裁判者）推进八阶段流程，并按需点名套件内两个计算器技能。" }, { "name": "labor-compensation-calculator", "description": "劳动补偿金与经济补偿金计算器（识律劳动法套件原子技能，双计算器之一）。当用户要求计算解除劳动合同经济补偿 N、违法解除赔偿金 2N、代通知金 N+1、未签书面劳动合同二倍工资差额、加班费、未休年休假工资报酬，或需要判断三倍封顶（社平工资三倍 × 12 年上限）时使用。法律依据走识律条级解析，数值基数走柏讯统计源，每个结果强制附公式与基数来源。" }, { "name": "personal-injury-calculator", "description": "人身损害赔偿计算器（识律劳动法套件原子技能，双计算器之一）。当用户要求计算人身损害赔偿项目时使用——医疗费、误工费、护理费、交通费、住院伙食补助费、营养费、残疾赔偿金、残疾辅助器具费、丧葬费、死亡赔偿金、被扶养人生活费、精神损害抚慰金。法律依据走识律条级解析，残疾/死亡赔偿金等基数走柏讯统计源，每个结果强制附公式与基数来源。" } ], "mcpServers": [ { "registryName": "fd-legal-search-mcp" }, { "registryName": "fd-find-data-business-mcp" } ] } }, { "id": "UzMQPvaZqdmpluQsbY-eng", "slug": "pack-uzmqpvazqdmpluqsby-eng", "category": "facade", "marketCategory": "legal", "tags": [ "识律", "法律", "lex" ], "canonical": { "brief": "Criminal defense suite: a master workflow chains plea/non-prosecution argumentation (review-for-prosecution) with examination outlines (first instance). Statutes and guiding cases via Lex legal search, validity via data-business search, similar cases via judgment search." }, "human": { "zh-CN": { "name": "识律 · 刑事辩护套件", "summary": "刑事辩护套件：以总控流程技能为入口，串联认罪认罚/不起诉论证（审查起诉阶段）与庭审发问提纲（一审阶段），法条与指导案例走识律法律检索、法规效力走数据业务检索，类案与量刑参照走裁判文书检索。" }, "en": { "name": "Lex · Criminal Defense Suite", "summary": "Criminal defense suite: a master workflow chains plea/non-prosecution argumentation (review-for-prosecution) with examination outlines (first instance). Statutes and guiding cases via Lex legal search, validity via data-business search, similar cases via judgment search." }, "es": { "name": "Lex · Suite de defensa penal", "summary": "Suite de defensa penal: un flujo maestro encadena clemencia por confesión / no procesamiento (revisión para procesar) con guiones de interrogatorio (primera instancia). Artículos y casos guía vía búsqueda jurídica Lex, vigencia vía datos de negocio, casos análogos vía sentencias." }, "fr": { "name": "Lex · Suite de défense pénale", "summary": "Suite de défense pénale : un flux maître enchaîne clémence pour aveu / non-poursuite (stade avant poursuites) et plans d'interrogatoire (première instance). Textes et affaires guide via la recherche Lex, validité via les données métier, affaires analogues via les décisions." }, "ja": { "name": "Lex · 刑事弁護スイート", "summary": "刑事弁護スイート：総括ワークフロースキルを入口に、認罪認罰・不起訴の論証（審査起訴段階）と法廷尋問提綱（第一審段階）を連携。法条と指導案例は Lex の法律検索、法規の効力はデータ業務検索、類似事案は裁判文書検索から。" } }, "whenToUse": "Browse under \"legal\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "criminal-defense-workflow", "description": "刑事辩护全流程总控（识律刑事辩护套件入口技能）。当用户提出刑事案件材料（起诉意见书/起诉书/卷宗/判决书）、要求刑事阅卷分析、辩点梳理、辩护策略路径选择、定性或量刑方向判断时使用。按审查起诉—一审的阶段推进（侦查阶段不在本技能主流程内，见阶段 1 说明），调度套件内「认罪认罚/不起诉论证」与「庭审发问提纲」两个原子技能，并统一承担法条锚定、类案检索与红线校验。" }, { "name": "criminal-plea-and-nonprosecution", "description": "认罪认罚与不起诉论证（识律刑事辩护套件原子技能）。审查起诉阶段使用：当用户问\"要不要认罪认罚\"\"认罪认罚协商\"\"量刑建议协商\"\"能不能不起诉\"\"法定/存疑/酌定不起诉\"或由 criminal-defense-workflow 调度时使用。输出证据态势、利弊权衡、量刑建议预测与协商三层条款、三种不起诉路径论证，并做当事人知情告知。" }, { "name": "criminal-trial-examination-outline", "description": "庭审发问与举证质证提纲（识律刑事辩护套件原子技能）。一审阶段使用：当用户说\"庭前会议\"\"排非申请\"\"非法证据排除\"\"庭审发问\"\"发问提纲\"\"质证提纲\"\"举证质证\"\"交叉询问\"\"逐份质证\"\"辩方举证\"或由 criminal-defense-workflow 调度时使用。输出庭前会议议程应对、非法证据排除申请、公诉人发问预判与应对、辩护人对被告人/证人/鉴定人的发问提纲、辩方举证提纲与逐份质证意见（书面版 + 庭上口述版）。" } ], "mcpServers": [ { "registryName": "fd-legal-search-mcp" }, { "registryName": "fd-find-data-business-mcp" } ] } }, { "id": "mfaIAXHRVFQJaXPXyiAzKQ", "slug": "pack-mfaiaxhrvfqjaxpxyiazkq", "category": "facade", "marketCategory": "legal", "tags": [ "识律", "法律", "lex" ], "canonical": { "brief": "Contract review under PRC (mainland) law: five industry review cards (labor & employment / loans & financing / NDA & confidentiality / sale & supply / services & SaaS) × statute verification with validity anchoring × law-bench clause assembly and triple review, producing a two-axis risk list, a revi" }, "human": { "zh-CN": { "name": "识律 · 合同审查套件", "summary": "中国法（大陆）合同审查：五张行业审查卡（劳动用工 / 借贷融资 / 保密NDA / 买卖供货 / 服务与SaaS）× 法条核验与效力锚定 × law-bench 条款装配与三重评审，产出双轴风险问题清单、审查备忘录与**修订对比表**（原条款/修订后条款/修订理由与核验依据三列，v2 新增交付形态）。" }, "en": { "name": "Lex · Contract Review Suite", "summary": "Contract review under PRC (mainland) law: five industry review cards (labor & employment / loans & financing / NDA & confidentiality / sale & supply / services & SaaS) × statute verification with validity anchoring × law-bench clause assembly and triple review, producing a two-axis risk list, a review memorandum and a revision comparison table (original clause / revised clause / rationale and verification basis — new in v2)." }, "es": { "name": "Lex · Suite de revisión de contratos", "summary": "Revisión de contratos en derecho continental chino: cinco fichas sectoriales (laboral / préstamos y financiación / NDA / compraventa y suministro / servicios y SaaS) × verificación de artículos con anclaje de vigencia × ensamblaje de cláusulas law-bench con triple revisión; produce una lista de riesgos de doble eje, un memorando de revisión y una tabla comparativa de revisiones (original / revisada / motivo y base de verificación — nuevo en v2)." }, "fr": { "name": "Lex · Suite de revue de contrats", "summary": "Revue de contrats en droit chinois (continental) : cinq fiches sectorielles (travail / prêts et financement / NDA / vente et fourniture / services et SaaS) × vérification des textes avec ancrage de validité × assemblage de clauses law-bench et triple revue ; produit une liste de risques à double axe, un mémorandum de revue et un tableau comparatif des révisions (clause d'origine / clause révisée / motif et base de vérification — nouveau en v2)." }, "ja": { "name": "Lex · 契約審査スイート", "summary": "中国法（大陸）での契約審査：5 枚の業界別審査カード（労務雇用／貸付融資／秘密保持 NDA／売買供給／サービス・SaaS）× 法条検証と効力アンカリング × law-bench による条項組み立てと三重レビュー。双軸リスク課題リスト・審査メモ・改訂対比表（原条項／改訂後条項／改訂理由と検証根拠の 3 列、v2 の新納品形態）を出力する。" } }, "whenToUse": "Browse under \"legal\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "contract-review-cn", "description": "中国法合同审查。判定合同适用中国大陆法后，加载通用原则与五张行业审查卡（劳动/借贷/保密/买卖/SaaS），逐条审查并产出双轴风险问题清单与审查备忘录；涉精确法条、阈值、生效日期时经法律数据工具核验。当用户要求审查/审阅/markup 合同、检查条款、问某条款在中国法下是否有效或如何修改时使用。仅覆盖中国大陆法，不含港澳台、不含刑事与行政、不含外国法规则包。" } ], "mcpServers": [ { "registryName": "fd-legal-search-mcp" }, { "registryName": "fd-find-data-business-mcp" }, { "registryName": "law-bench" } ] } }, { "id": "rz-ebXHD5astVGcyqlhmqA", "slug": "pack-rz-ebxhd5astvgcyqlhmqa", "category": "facade", "marketCategory": "legal", "tags": [ "识律", "法律", "lex" ], "canonical": { "brief": "End-to-end analysis of civil and commercial cases: fixing the claims (S1) → basis of the claim (S2) → defense analysis (S3) → element breakdown (S4) → assertion search (S5) → issue matrix (S6) → allocation of the burden of proof (S7) → fact-finding (S8) → outcome prediction (S9) → hallucination chec" }, "human": { "zh-CN": { "name": "识律 · 民事诉讼套件", "summary": "民商事案件全流程分析：固定诉讼请求（S1）→ 请求权基础（S2）→ 抗辩分析（S3）→ 要件拆解（S4）→ 主张检索（S5）→ 争点矩阵（S6）→ 举证责任分配（S7）→ 事实认定（S8）→ 裁判预测（S9）→ 幻觉检查（S10）。法条走识律语义检索与条级解析、效力状态走数据业务检索、类案与指导案例走裁判文书与指导案例库；S10 由原自检改为**用数据工具反查**引用真实性。" }, "en": { "name": "Lex · Civil Litigation Suite", "summary": "End-to-end analysis of civil and commercial cases: fixing the claims (S1) → basis of the claim (S2) → defense analysis (S3) → element breakdown (S4) → assertion search (S5) → issue matrix (S6) → allocation of the burden of proof (S7) → fact-finding (S8) → outcome prediction (S9) → hallucination check (S10). Statutes via Lex semantic search with article-level parsing, validity status via data-business search, similar and guiding cases via the judgment and guiding-case libraries; S10 now verifies citation authenticity by cross-checking with the data tools." }, "es": { "name": "Lex · Suite de litigación civil", "summary": "Análisis completo de casos civiles y mercantiles: fijación de pretensiones (S1) → base de la pretensión (S2) → análisis de defensas (S3) → desglose de elementos (S4) → búsqueda de afirmaciones (S5) → matriz de cuestiones (S6) → carga de la prueba (S7) → apreciación de los hechos (S8) → predicción de la resolución (S9) → control de alucinaciones (S10). Artículos vía búsqueda semántica de Lex, vigencia vía datos de negocio, casos análogos y guía vía las bases de sentencias y casos guía; S10 verifica las citas con las herramientas de datos." }, "fr": { "name": "Lex · Suite de contentieux civil", "summary": "Analyse complète des affaires civiles et commerciales : fixation des prétentions (S1) → fondement de la demande (S2) → analyse des défenses (S3) → décomposition des éléments (S4) → recherche des allégations (S5) → matrice des points en litige (S6) → charge de la preuve (S7) → appréciation des faits (S8) → prévision de la décision (S9) → contrôle des hallucinations (S10). Textes via la recherche sémantique Lex, validité via les données métier, affaires analogues et guide via les bases de décisions ; S10 vérifie les citations avec les outils de données." }, "ja": { "name": "Lex · 民事訴訟スイート", "summary": "民商事事件の全工程分析：訴訟請求の確定（S1）→ 請求権基礎（S2）→ 抗弁分析（S3）→ 要件分解（S4）→ 主張検索（S5）→ 争点マトリクス（S6）→ 証明責任の配分（S7）→ 事実認定（S8）→ 裁判結果予測（S9）→ ハルシネーション検査（S10）。法条は Lex のセマンティック検索と条文単位の解析、効力状態はデータ業務検索、類似事案・指導案例は裁判文書庫と指導案例庫から。S10 は従来の自己点検から、データツールによる引用の真実性照合に変更。" } }, "whenToUse": "Browse under \"legal\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "civil-nine-step-workflow", "description": "民事案件九步法会话版（请求权基础 + 要件审判九步法 + 规范说）。当用户描述案情或提供起诉状/答辩状/证据/判决书，要求固定诉讼请求、找请求权基础、抗辩分析、要件拆解、类案检索、争点矩阵、举证责任分配、事实认定、裁判预测或法条/案号真实性核验时使用。输出为纯文本会话推演，不写文件系统、不调脚本。" } ], "mcpServers": [ { "registryName": "fd-legal-search-mcp" }, { "registryName": "fd-find-data-business-mcp" } ] } }, { "id": "3LUWFBGS3RJe5BIV9MvBFw", "slug": "pack-3luwfbgs3rje5biv9mvbfw", "category": "pack", "marketCategory": "facet", "tags": [ "官方功能集", "指南", "平台" ], "canonical": { "brief": "Operation and authoring guides for the Finddata platform: 1) the Yizuo and Wanxing manuals — using the AI assistant workbench, installing packs, deploying and calling agent services (A2A); 2) an orchestration supplement to Facet pack authoring — the batch-suite definition schema, registry-side skill" }, "human": { "zh-CN": { "name": "平台指南", "summary": "寻数平台操作与制作指南：①壹座与萬星使用手册——AI 助手工作台怎么用、功能集怎么装、Agent 服务怎么部署与调用（A2A）；②谦面功能集制作的**编排补篇**——批量套件的定义 schema、registry 侧技能注册、权限与计费四道闸全貌。（**想自己做一个功能集，请装《谦面工坊》**：它带发现与发布工具、覆盖完整创作流程；本包的制作技能只讲工坊不覆盖的平台内部细节。）平台官方维护，随平台演进更新。" }, "en": { "name": "Platform Guides", "summary": "Operation and authoring guides for the Finddata platform: 1) the Yizuo and Wanxing manuals — using the AI assistant workbench, installing packs, deploying and calling agent services (A2A); 2) an orchestration supplement to Facet pack authoring — the batch-suite definition schema, registry-side skill registration, and the full picture of the four gates for permissions and billing. To author a pack yourself, install Facet Forge: it covers the whole creation flow; this pack's skills only cover platform internals the Forge does not. Maintained by the platform, updated as it evolves." }, "es": { "name": "Guías de la plataforma", "summary": "Guías de operación y creación de la plataforma Finddata: 1) los manuales de Yizuo y Wanxing — usar el banco de trabajo del asistente de IA, instalar packs, desplegar y llamar servicios de agentes (A2A); 2) un suplemento de orquestación para la creación de packs en Facet — el esquema de suites por lotes, el registro de habilidades en el registry y las cuatro puertas de permisos y facturación. Para crear un pack tú mismo, instala Facet Forge: cubre todo el flujo; este pack trata solo los detalles internos que la Forge no ve. Mantenida por la plataforma." }, "fr": { "name": "Guides de la plateforme", "summary": "Guides d'utilisation et de création de la plateforme Finddata : 1) les manuels de Yizuo et Wanxing — utiliser l'établi de l'assistant IA, installer des packs, déployer et appeler des services d'agents (A2A) ; 2) un supplément d'orchestration à la création de packs Facet — le schéma des suites par lot, l'enregistrement des compétences côté registry et les quatre portes de permissions et de facturation. Pour créer un pack vous-même, installez Facet Forge : il couvre tout le flux ; ce pack ne traite que les détails internes que la Forge ne voit pas. Maintenu par la plateforme." }, "ja": { "name": "プラットフォームガイド", "summary": "Finddata プラットフォームの操作・制作ガイド：① 壹座と萬星の取扱説明書 — AI アシスタント作業台の使い方、機能セットの導入、Agent サービスのデプロイと呼び出し（A2A）；② Facet 機能セット制作の編成サプリ — バッチスイートの定義スキーマ、registry 側のスキル登録、権限と課金の四つのゲートの全体像。（自分で機能セットを作りたいなら『Facet フォージ』を導入してください：発見と公開のツールを備え、制作フロー全体をカバーします。本パックの制作技能は、フォージが扱わないプラットフォーム内部の詳細のみ。）プラットフォーム公式が維持し、進化に合わせて更新。" } }, "whenToUse": "Browse under \"facet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "yizuo-wanxing-guide", "description": "壹座与萬星使用手册：壹座（AI 助手工作台）的界面地图、功能集安装动线、自建预设、常见操作；萬星（Agent 服务部署与运行平台）的服务目录、A2A 调用、部署者控制台、访客对话入口；含两平台分工关系与五分钟客户演示动线。当用户询问如何操作这两个平台、如何部署或调用 agent 服务、如何向客户演示时使用。" }, { "name": "facet-pack-builder", "description": "谦面功能集（pack）制作的内部视角补篇：批量套件编排（套件定义 schema）、registry 侧技能注册、权限与计费四道闸的全貌。单包/首次制作请用《谦面工坊》功能集（它带发现与发布工具、覆盖完整创作流程）；本技能只讲工坊不覆盖的编排层与平台内部细节。当用户要批量生产一整条产品线的套件、要把技能注册进 registry 供各处复用、或要排查\"装不上/用不了\"的四道闸时使用。" } ], "mcpServers": [] } }, { "id": "_B99vDq0CFBNQGvmfy5iSw", "slug": "pack-b99vdq0cfbnqgvmfy5isw", "category": "facade", "marketCategory": "facet", "tags": [ "官方功能集", "创作", "指南" ], "canonical": { "brief": "The pack-forging workshop for ecosystem creators: turn an idea into an installable pack. 1) Forge — discover the platform's MCP tools and skills, check for duplicates, assemble the manifest, pre-validate, publish. 2) Adapt — license review and attribution when building on open-source or external ski" }, "human": { "zh-CN": { "name": "谦面工坊", "summary": "生态创作者的铸包工坊：把一个想法做成可安装的功能集（pack）。①铸包——发现平台可用的 MCP 工具与技能、查重、组装 manifest、预校验、发布；②改编——用开源或外部技能打底时的许可证判定与署名（含降级路径与防编造约束）；③验收——发布后自查：装得上吗、工具真的被调用了吗、取不到时说得诚实吗。面向生态创作者；发布需创作者身份。" }, "en": { "name": "Facet Forge", "summary": "The pack-forging workshop for ecosystem creators: turn an idea into an installable pack. 1) Forge — discover the platform's MCP tools and skills, check for duplicates, assemble the manifest, pre-validate, publish. 2) Adapt — license review and attribution when building on open-source or external skills, with fallback paths and no-fabrication constraints. 3) Accept — post-publish self-check: does it install, are tools really called, honest when data is missing. Publishing requires creator status." }, "es": { "name": "Forja de Facet", "summary": "Taller de forja de packs para creadores del ecosistema: convierte una idea en un pack instalable. 1) Forjar — descubre herramientas MCP y habilidades, comprueba duplicados, arma el manifest, prevalida y publica. 2) Adaptar — licencias y atribución al basarse en habilidades de código abierto o externas, con degradación honesta y sin inventar. 3) Aceptar — autochequeo tras publicar: ¿se instala?, ¿las herramientas se invocan de verdad?, ¿es honesto si faltan datos? Publicar exige condición de creador." }, "fr": { "name": "La Forge Facet", "summary": "Atelier de forge à packs pour les créateurs de l'écosystème : transformer une idée en pack installable. 1) Forger — découvrir les outils MCP et compétences, vérifier les doublons, assembler le manifest, prévalider, publier. 2) Adapter — licence et attribution quand on part de compétences open source ou externes, avec repli honnête. 3) Réceptionner — autocontrôle après publication : installation, appels réels des outils, honnêteté si les données manquent. Publier exige le statut créateur." }, "ja": { "name": "Facet フォージ", "summary": "エコシステムクリエイターのための鋳包工房：アイデアをインストール可能な機能セット（pack）に仕上げます。①鋳造 — プラットフォームで使える MCP ツールとスキルを探索し、重複を確認し、manifest を組み立て、事前検証を経て公開。②改変 — オープンソースや外部スキルを下敷きにするときのライセンス判定と帰属表示（フォールバック経路と捏造禁止の制約つき）。③検収 — 公開後のセルフチェック：インストールできるか、ツールは本当に呼ばれたか、取れないときに誠実か。エコシステムクリエイター向け。公開にはクリエイター資格が必要です。" } }, "whenToUse": "Browse under \"facet\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "facet-forge-assembly", "description": "在谦面把想法铸成功能集（pack）：先发现可用的 MCP 工具与技能，再组装 manifest（内联技能 + MCP 引用 + 角色），预校验通过后发布。用户要「做一个功能集/预设套件/上架」「把某能力打包给用户装」「我有个想法想做成 AI 助手」时使用。" }, { "name": "facet-forge-adaptation", "description": "把别人的技能（GitHub 开源技能、外部技能市场条目、谦面已上架 pack 里的技能）改编成自己的功能集内容时的合规与改编方法：许可证能不能用、怎么署名、怎么把数据环节换成平台工具。用户说\"照着这个开源项目做一个\"\"把这个技能抄过来改改\"\"参考某某套件\"时使用。" }, { "name": "facet-forge-acceptance", "description": "功能集发布之后的自查清单：确认用户装得上、装上了能用、工具真的被调用过、工具说做不到时是否诚实。用户说\"我发布了这个功能集，接下来呢\"\"怎么知道它有没有用\"\"用户反馈装不上/用不了\"时使用。" } ], "mcpServers": [ { "registryName": "airegistry-tools" }, { "registryName": "facet-market-mcp" } ] } }, { "id": "XfyXvzONQFelnFIgM0dJVw", "slug": "pack-xfyxvzonqfelnfigm0djvw", "category": "facade", "marketCategory": "legal", "tags": [ "lex", "copilot", "finddata" ], "canonical": { "brief": "A copilot for the Lex console: a five-page tour, reading quota and billing methodology, credential guidance; after an ask-card confirmation it can propose navigate/refresh page actions (executed by the host page). No account-data read/write tools, and never fabricates numbers." }, "human": { "zh-CN": { "name": "识律控制台伴随助手", "summary": "识律控制台的副驾：五页导览、配额与账单口径解读、凭据管理指引；经问询卡确认后可提议 navigate/refresh 页面动作（由宿主页面执行）。无账户数据读取与写操作工具，绝不编造数字。" }, "en": { "name": "Lex Console Companion", "summary": "A copilot for the Lex console: a five-page tour, reading quota and billing methodology, credential guidance; after an ask-card confirmation it can propose navigate/refresh page actions (executed by the host page). No account-data read/write tools, and never fabricates numbers." }, "es": { "name": "Compañero de la consola Lex", "summary": "Copiloto de la consola Lex: recorrido de cinco páginas, lectura de cuotas y facturación, guía de credenciales; tras confirmar propone acciones navigate/refresh (las ejecuta la página anfitriona). Sin herramientas de datos de cuenta y sin inventar cifras." }, "fr": { "name": "Compagnon de la console Lex", "summary": "Copilote de la console Lex : visite en cinq pages, lecture des quotas et de la facturation, guide des identifiants ; après confirmation, il propose des actions navigate/refresh (exécutées par la page hôte). Aucun outil sur les données de compte, jamais de chiffres inventés." }, "ja": { "name": "Lex コンソールコンパニオン", "summary": "識律コンソールのコパイロット：5 ページの案内、割り当てと請求の算出基準の読み方、認証情報の管理ガイド；確認カードの承認のあとで navigate/refresh のページ操作を提案できます（実行はホストページ）。アカウントデータの読み書きツールは持たず、数字を捏造しません。" } }, "whenToUse": "Browse under \"legal\" in the market.", "badges": { "official": false, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "lex-console-tour", "description": "识律控制台五页导览与路由表——每页管什么、用户问题该去哪页、何时发起 navigate 意图" }, { "name": "lex-credentials-guide", "description": "识律凭据管理指引——创建、轮换、泄露处置的通用流程与安全纪律；不代用户执行" }, { "name": "lex-intent-protocol", "description": "宿主页面动作协议——确认纪律、被拒后的改口规则、什么算动作什么不算；信封的字面形状见 persona" }, { "name": "lex-quota-billing", "description": "识律配额与账单的解读方法——口径以页面为准、硬停恢复途径、常见问题话术，绝不编造数字" } ], "mcpServers": [] } }, { "id": "MnViPe6mi-FBOUL1Z0tDTw", "slug": "pack-mnvipe6mi-fboul1z0tdtw", "category": "pack", "marketCategory": "fleet", "tags": [], "canonical": { "brief": "选题→初稿→审校的流水线，只出草稿供人定稿；不自动发布" }, "human": { "zh-CN": { "name": "内容工厂", "summary": "选题→初稿→审校的流水线，只出草稿供人定稿；不自动发布" }, "en": { "name": "内容工厂", "summary": "选题→初稿→审校的流水线，只出草稿供人定稿；不自动发布" }, "es": { "name": "内容工厂", "summary": "选题→初稿→审校的流水线，只出草稿供人定稿；不自动发布" }, "fr": { "name": "内容工厂", "summary": "选题→初稿→审校的流水线，只出草稿供人定稿；不自动发布" }, "ja": { "name": "内容工厂", "summary": "选题→初稿→审校的流水线，只出草稿供人定稿；不自动发布" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [], "mcpServers": [] } }, { "id": "bKipx4urGfO5TE6GjJ9Ihg", "slug": "pack-bkipx4urgfo5te6gjj9ihg", "category": "pack", "marketCategory": "fleet", "tags": [ "团队", "基础设施", "巡检", "优化", "报班" ], "canonical": { "brief": "Owner-facing three-post fleet: the sentinel audits all registered fleets and infrastructure against countable thresholds and lands a per-item anomaly ledger; the optimizer routes confirmed anomalies through existing proposal surfaces only, each citing its ledger entry; the shift lead delivers a week" }, "human": { "zh-CN": { "name": "工厂集群", "summary": "工厂集群：Owner 自用的三岗运行编制——巡检岗按可计数阈值观测全部在册集群与基础设施并逐条目落异常清单；优化岗把已确认异常只经既有提案面转为配置/蓝图/技能/跨面提案并附条目依据；报班岗按周汇总异常台账、提案四态计数与成本走势，拍板件走 decision_queue 正式呈报。提案与批准分离：红线动作永远人工，批准前零写动作。 含集群模板 fleet 段（三岗编排：巡检→优化→报班；装包后经壹座「团队→从模板实例化」建队）。" }, "en": { "name": "Infra Factory Fleet", "summary": "Owner-facing three-post fleet: the sentinel audits all registered fleets and infrastructure against countable thresholds and lands a per-item anomaly ledger; the optimizer routes confirmed anomalies through existing proposal surfaces only, each citing its ledger entry; the shift lead delivers a weekly digest and puts decisions in the decision queue. Proposals and approvals stay separate; red-line actions are always human." }, "es": { "name": "Clúster Fábrica", "summary": "Clúster de tres puestos para el propietario: el centinela audita todos los clústeres registrados y la infraestructura contra umbrales contables y deja un registro de anomalías por elemento; el optimizador convierte las anomalías confirmadas solo por las superficies de propuesta existentes, citando su entrada del registro; el jefe de turno entrega un informe semanal y presenta las decisiones en la cola. Propuestas y aprobaciones se mantienen separadas; las acciones de línea roja siempre son humanas." }, "fr": { "name": "Cluster Usine", "summary": "Cluster à trois postes pour le propriétaire : la sentinelle audite tous les clusters enregistrés et l'infrastructure selon des seuils comptables et tient un registre d'anomalies par élément ; l'optimiseur transforme les anomalies confirmées uniquement via les surfaces de proposition existantes, en citant l'entrée du registre ; le chef de quart livre un digest hebdomadaire et soumet les décisions dans la file. Propositions et validations restent séparées ; les actions de ligne rouge sont toujours humaines." }, "ja": { "name": "工場クラスタ", "summary": "工場クラスタ：Owner 自身が使う3役の運行編成——巡回役は計数可能なしきい値で全在冊クラスタと基盤を観測し、項目別の異常台帳を残す；最適化役は確認済みの異常を既存の提案面のみで設定／ブループリント／スキル／横断提案に変換し、台帳項目を根拠として添える；報告役は週次で異常台帳・提案四状態の計数・コスト推移をまとめ、承認事項は decision_queue で正式に提出する。提案と承認は分離：レッドライン動作は常に人手、承認前は書き込みゼロ。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "sentinel-discipline", "description": "工厂集群巡检岗纪律：五类可计数阈值判据、逐条目异常清单格式、单回合口径收窄与收尾报数落 board。装配于工厂集群巡检岗。" }, { "name": "optimizer-discipline", "description": "工厂集群优化岗纪律：异常→提案四通道映射、提案必附异常条目依据、幻觉源名防御、自动化政策线分级标注、批准前零写边界。装配于工厂集群优化岗。" }, { "name": "shiftlead-discipline", "description": "工厂集群报班岗纪律：周报 digest 三段（异常台账/提案四态计数/成本走势）、拍板件走 decision_queue 正式呈报、禁私信替代。装配于工厂集群报班岗。" } ], "mcpServers": [ { "registryName": "infra-telemetry" }, { "registryName": "fleet-architect" }, { "registryName": "blueprint-author" }, { "registryName": "facet-market" }, { "registryName": "actions-mcp" } ] } }, { "id": "C8wx14-4h7zJ0N19-J4NwQ", "slug": "pack-c8wx14-4h7zj0n19-j4nwq", "category": "pack", "marketCategory": "fleet", "tags": [ "团队", "跨境电商", "政策" ], "canonical": { "brief": "A policy-watch cluster over customs/commerce/tax/port announcement lanes, daily brief with three escalation criteria, board draft contribution. Built from a natural-language brief by the vertical constructor." }, "human": { "zh-CN": { "name": "跨境电商政策盯梢团队", "summary": "跨境电商政策面值守团队：监视位（海关/商务部/税务/口岸四路 fd-info 公告流）→合成位（政策日报+三判据 escalate）→产出位（看板草稿，发布走提案）。纪律来自共用层智能体（同 cell 装）；由蓝图构造间按自然语言需求建成，可溯源。" }, "en": { "name": "Cross-border E-commerce Policy Watch", "summary": "A policy-watch cluster over customs/commerce/tax/port announcement lanes, daily brief with three escalation criteria, board draft contribution. Built from a natural-language brief by the vertical constructor." }, "es": { "name": "Vigilancia de Políticas de Comercio Electrónico Transfronterizo", "summary": "Clúster de vigilancia de políticas sobre los canales de anuncios de aduanas, comercio, impuestos y puertos: informe diario con tres criterios de escalado y borradores para el tablero. Construido a partir de una solicitud en lenguaje natural por el constructor vertical." }, "fr": { "name": "Veille des politiques de commerce électronique transfrontalier", "summary": "Cluster de veille des politiques sur les canaux d'annonces douanes, commerce, fiscalité et ports : brief quotidien avec trois critères d'escalade et brouillons pour le tableau de bord. Construit à partir d'une demande en langage naturel par le constructeur vertical." }, "ja": { "name": "越境EC政策ウォッチ", "summary": "税関・商務・税務・通関の4系統の公告を監視し、日次ブリーフと3つのエスカレーション基準で判断、掲示板ドラフトを提出する政策監視クラスタ。自然言語の依頼から垂直コンストラクタが構築し、来歴を追跡できる。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "crossborder-four-lane-patrol", "description": "四路政策巡检配方：海关/商务/税务/口岸各自的检索式、水位与原文优先级。" }, { "name": "crossborder-regime-code-catalog", "description": "监管代码目录：9610/1210/9710/9810 等代码的适用场景与单列纪律。" }, { "name": "crossborder-policy-conflict-triage", "description": "三判据分流：新征/调口径、与昨日冲突、多源不一致，命中即 high escalate。" } ], "mcpServers": [ { "registryName": "fd-info" }, { "registryName": "wire-ops" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "8wJVZZnk369wm0oFScdecA", "slug": "pack-8wjvzznk369wm0ofscdeca", "category": "pack", "marketCategory": "fleet", "tags": [ "团队", "运营" ], "canonical": { "brief": "Wanxing fleet-ops fleet: health patrol over fleets.db (read-only, names failing members), agent steward over the wanxing read face (degrades honestly without credentials), and an ops chief that composes the daily brief and escalates governance proposals." }, "human": { "zh-CN": { "name": "萬星机群运营团队", "summary": "萬星团队运行面运营团队：监视位（fleets.db 只读直查，连续失败点名到 member）+归账位（wanxing 只读盘点，凭据缺位如实降级）+合成位（每日简报+治理提案）。治理写操作一律提案+拍板（allowTools 工具面级排除）。" }, "en": { "name": "Wanxing Fleet Ops", "summary": "Wanxing fleet-ops fleet: health patrol over fleets.db (read-only, names failing members), agent steward over the wanxing read face (degrades honestly without credentials), and an ops chief that composes the daily brief and escalates governance proposals." }, "es": { "name": "Operaciones de Flota Wanxing", "summary": "Clúster de operaciones del plano de equipos Wanxing: monitor (fleets.db en solo lectura, nombra los miembros que fallan), registrador (inventario wanxing en solo lectura, degradación honesta sin credenciales) y sintetizador (informe diario y propuestas de gobernanza). Las escrituras de gobernanza siempre por propuesta y aprobación." }, "fr": { "name": "Opérations de flotte Wanxing", "summary": "Cluster d'exploitation du plan d'équipes Wanxing : moniteur (fleets.db en lecture seule, nomme les membres en échec), registraire (inventaire wanxing en lecture seule, dégradation honnête sans identifiants) et synthétiseur (brief quotidien et propositions de gouvernance). Les écritures de gouvernance passent toujours par proposition et validation." }, "ja": { "name": "萬星機群運用チーム", "summary": "萬星チーム運用面の運営クラスタ：監視役（fleets.db を読み取り専用で直接照会し、連続失敗を member 単位で指名）＋集計役（wanxing の読み取り専用棚卸、資格情報が欠ける場合は正直に降格）＋統合役（日次ブリーフとガバナンス提案）。ガバナンスの書き込みは常に提案＋承認（allowTools で書き込みを除外）。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "fleet-readonly-db-patrol", "description": "fleets.db 只读巡检配方：表清单、成员态枚举、连续失败计数口径与 vacant 时长算法。" }, { "name": "fleet-governance-proposal-card", "description": "治理提案卡：现象/证据/选项/影响面四栏，写面动作一律提案不自决。" } ], "mcpServers": [ { "registryName": "wanxing-ops" } ] } }, { "id": "ujUI1A2caCUW7TG1Z1mlEg", "slug": "pack-ujui1a2cacuw7tg1z1mleg", "category": "pack", "marketCategory": "fleet", "tags": [ "团队", "运营" ], "canonical": { "brief": "Facet platform-ops cluster: inventory patrol (read-only), daily integrity sentinel that degrades honestly when integrity is not exposed, and an ops chief that composes the brief and escalates curation proposals. Read-only tools only — curation is proposal-gated." }, "human": { "zh-CN": { "name": "谦面运营团队", "summary": "谦面市场生态运营团队：监视位（周一深度盘点）+闸位（每日完整性日检，integrity 不可得如实降级）+合成位（每日简报+策写提案）。只读面巡检，策写一律提案+拍板（allowTools 工具面级排除）。" }, "en": { "name": "Facet Platform Ops", "summary": "Facet platform-ops cluster: inventory patrol (read-only), daily integrity sentinel that degrades honestly when integrity is not exposed, and an ops chief that composes the brief and escalates curation proposals. Read-only tools only — curation is proposal-gated." }, "es": { "name": "Operaciones de Plataforma Facet", "summary": "Clúster de operaciones del ecosistema de Facet: monitor (inventario semanal), guardián (verificación diaria de integridad, con degradación honesta si no está disponible) y sintetizador (informe diario y propuestas de curación). Patrulla de solo lectura; la curación siempre pasa por propuesta y aprobación." }, "fr": { "name": "Opérations de la plateforme Facet", "summary": "Cluster d'exploitation Facet : moniteur (inventaire hebdo), gardien (contrôle d'intégrité quotidien, dégradation honnête si indisponible) et synthétiseur (brief quotidien + propositions de curation). Patrouille en lecture seule ; curation toujours sur proposition et validation." }, "ja": { "name": "Facet 運営チーム", "summary": "Facet 市場エコシステムの運営クラスタ：監視役（週次の棚卸）＋ゲート役（毎日の整合性検査、取得不能時は正直に降格）＋統合役（日次ブリーフとキュレーション提案）。巡回は読み取り専用、キュレーションは常に提案＋承認（allowTools で書き込みを除外）。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "facet-visibility-inventory", "description": "可见性盘点口径：按档位分计、同条重复标注、口径字段固定，保证跨轮可比。" }, { "name": "facet-integrity-evidence-ladder", "description": "完整性证据阶梯：字段存在→字段可比→字段不一致三级，未暴露只报未取得。" } ], "mcpServers": [ { "registryName": "facet-market" } ] } }, { "id": "LjTJZWqSr2AOX7WEFTVlpg", "slug": "pack-ljtjzwqsr2aox7weftvlpg", "category": "pack", "marketCategory": "fleet", "tags": [ "团队", "营销", "官网" ], "canonical": { "brief": "Marketing content fleet template: topics → draft → review → publish-draft chain (staggered crons) plus an eval sentinel post. Magnitude-only wording, review is the single gate, publishing goes through the proposal channel. Post discipline comes from the shared fleet packs." }, "human": { "zh-CN": { "name": "官网内容团队", "summary": "营销线团队模板：选题→撰稿→审读→发布草稿四岗 handoff 链（错峰 cron ≥30min）+闸位岗独立对拍（gold 命中/覆盖，评测报告落草稿）。量级话术红线（禁精确数字）、审读放行唯一门、发布走提案不直发。装包后经壹座「从模板实例化」建队；纪律来自共用层四包（同 cell 装）。" }, "en": { "name": "Marketing Site Content", "summary": "Marketing content fleet template: topics → draft → review → publish-draft chain (staggered crons) plus an eval sentinel post. Magnitude-only wording, review is the single gate, publishing goes through the proposal channel. Post discipline comes from the shared fleet packs." }, "es": { "name": "Contenido del Sitio de Marketing", "summary": "Plantilla de clúster de marketing: cadena de cuatro roles temas → redacción → revisión → borrador de publicación (crons escalonados ≥30 min) más un puesto de evaluación independiente. Lenguaje de magnitud (sin cifras precisas), la revisión es la única compuerta, la publicación pasa por propuesta y nunca es directa. Instala y crea la instancia desde la plantilla; la disciplina viene de los cuatro packs compartidos (misma celda)." }, "fr": { "name": "Contenu du site marketing", "summary": "Modèle de cluster marketing : chaîne de quatre rôles sujets → rédaction → relecture → brouillon de publication (crons décalés ≥30 min) plus un poste d'évaluation indépendant. Formulations de magnitude (sans chiffres précis), la relecture est l'unique porte, la publication passe par proposition et jamais en direct. Installez puis instanciez depuis le modèle ; la discipline vient des quatre packs partagés (même cellule)." }, "ja": { "name": "マーケティングサイト・コンテンツ", "summary": "マーケティング線のクラスタテンプレート：テーマ選定→執筆→レビュー→公開ドラフトの4役ハンドオフ（cron を30分以上ずらす）＋独立した評価役。桁感の表現を守り（正確な数字は禁止）、レビューが唯一のゲート、公開は提案経由で直接公開しない。インストール後にテンプレートからインスタンス化。規律は共用4パック（同じセルに導入）から。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "web-assertion-source-ledger", "description": "断言溯源台账：数字/因果/承诺三类断言逐条落来源，写不出来源就降级或标待验证。" }, { "name": "web-machine-readable-sections", "description": "机器可读五条自查：小节自足、术语唯一、结论先行、表格优先、标题即断言锚点。" }, { "name": "web-cadence-brief-contract", "description": "四岗 handoff 契约：选题→成稿→放行→落稿的交接物清单与错峰门槛。" } ], "mcpServers": [ { "registryName": "fd-info" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "IVCGxpg9itGnLpHxT1khWQ", "slug": "pack-ivcgxpg9itgnlphxt1khwq", "category": "pack", "marketCategory": "fleet", "tags": [ "团队", "柏讯", "情报" ], "canonical": { "brief": "Wire intel/sentiment fleet template: intel patrol + sentiment sentinel run in parallel, the chief cross-checks both and composes the morning brief, the contribution post lands a board draft. Install, then instantiate from the template; post discipline comes from the shared fleet packs (install in th" }, "human": { "zh-CN": { "name": "情报舆情团队", "summary": "柏讯情报舆情团队模板：监视位→闸位双路并行→合成位交叉核对→产出位落看板草稿（trigger×3+handoff×3 编导）。装包后经壹座「团队→从模板实例化」建队；岗位纪律来自共用层四包（需同 cell 安装）：fleet-intel-gather/fleet-data-qa/fleet-brief-report/fleet-board-contribution。" }, "en": { "name": "Wire Intel & Sentiment", "summary": "Wire intel/sentiment fleet template: intel patrol + sentiment sentinel run in parallel, the chief cross-checks both and composes the morning brief, the contribution post lands a board draft. Install, then instantiate from the template; post discipline comes from the shared fleet packs (install in the same cell)." }, "es": { "name": "Inteligencia y Sentimiento de Wire", "summary": "Plantilla de clúster de inteligencia y sentimiento de Wire: el monitor y el guardián de sentimiento trabajan en paralelo, el sintetizador cruza ambos y compone el informe matinal, y el productor guarda un borrador para el tablero (3 trigger + 3 handoff). Instala y crea la instancia desde la plantilla; la disciplina viene de los cuatro packs compartidos (misma celda): fleet-intel-gather / fleet-data-qa / fleet-brief-report / fleet-board-contribution." }, "fr": { "name": "Renseignement et sentiment Wire", "summary": "Modèle de cluster renseignement et sentiment Wire : le moniteur et le gardien de sentiment travaillent en parallèle, le synthétiseur recoupe les deux et compose le brief du matin, le producteur dépose un brouillon pour le tableau de bord (3 trigger + 3 handoff). Installez puis instanciez depuis le modèle ; la discipline vient des quatre packs partagés (même cellule) : fleet-intel-gather / fleet-data-qa / fleet-brief-report / fleet-board-contribution." }, "ja": { "name": "Wire 情報・センチメント", "summary": "Wire 情報・センチメントのクラスタテンプレート：監視役とセンチメントゲート役が並行し、統合役が両者をクロスチェックして朝のブリーフを作成、制作役が掲示板の下書きを保存（trigger×3 + handoff×3 の編成）。インストール後にテンプレートからインスタンス化。規律は共用4パック（同じセルに導入）：fleet-intel-gather / fleet-data-qa / fleet-brief-report / fleet-board-contribution。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "intel-item-five-elements", "description": "情报条目五要素：原文标题、时间、来源、实体、可回溯 ID，缺一不立条。" }, { "name": "intel-source-health-triage", "description": "源健康先行：stream_health 判红黄绿三档，坏源不硬取，失败按四分类上报。" }, { "name": "intel-sentiment-threshold-card", "description": "情绪阈值卡：基线算法、窗宽、阈值来源与连续同向判据的可复核口径。" } ], "mcpServers": [ { "registryName": "fd-info" }, { "registryName": "wire-ops" }, { "registryName": "wire-build-mcp" }, { "registryName": "fd-sentiment" } ] } }, { "id": "YT87-2Axuu49I123zVYyhA", "slug": "pack-yt87-2axuu49i123zvyyha", "category": "pack", "marketCategory": "fleet", "tags": [ "团队", "柏讯", "本体" ], "canonical": { "brief": "A Wire ontology-building cluster for the data plane: profile datasets, draft entity/relation declarations, pass deterministic validation, and file declarations through the customer action gate so the entity graph holds real content. Five-role pipeline (profiler → modeler → deterministic gate → propo" }, "human": { "zh-CN": { "name": "柏讯本体建设团队", "summary": "柏讯数据本体建设团队：对靶客户数据面持续盘账、建模、校验、经动作闸提议落库，让实体关系图谱有真实内容。五岗流水线：盘账员（只读画像：行列规模、候选主键、疑似外键）→建模师（实体/关系声明草案，业务名从列名直译禁臆造）→纪律校验闸（确定性规则放行或拦截，给列名级原因）→声明提案员（经 wire_action_propose 提议内置声明动作，即时返回即合规，写操作永不直调）→台账主编（三点实调抽样对账，缺位如实标注）。全链只读+提案制：本体声明经客户动作闸确认后落库，数据行零写回，实体与关系只建不改不删。" }, "en": { "name": "Wire Ontology Builder", "summary": "A Wire ontology-building cluster for the data plane: profile datasets, draft entity/relation declarations, pass deterministic validation, and file declarations through the customer action gate so the entity graph holds real content. Five-role pipeline (profiler → modeler → deterministic gate → proposal clerk → ledger editor) with read-only discipline, zero data-row writes, create-only v0." }, "es": { "name": "Constructor de Ontología de Wire", "summary": "Clúster de construcción de ontología para el plano de datos de Wire: perfila conjuntos de datos, redacta declaraciones de entidades y relaciones, pasa la validación determinista y presenta las declaraciones por la compuerta de acciones del cliente para que el grafo tenga contenido real. Pipeline de cinco roles (perfilador → modelador → compuerta determinista → proponente → editor del registro), solo lectura y cero escrituras de filas." }, "fr": { "name": "Constructeur d'ontologie Wire", "summary": "Cluster de construction d'ontologie pour le plan de données Wire : profiler les jeux de données, rédiger les déclarations d'entités et de relations, passer la validation déterministe et déposer les déclarations via la porte d'actions du client afin que le graphe contienne du contenu réel. Pipeline à cinq rôles (profileur → modeleur → porte déterministe → chargé des déclarations → rédacteur du registre), lecture seule et zéro écriture de lignes." }, "ja": { "name": "Wire オントロジー構築チーム", "summary": "Wire データ面のオントロジー構築クラスタ：データセットを棚卸しし、エンティティ／関係の宣言草案を作成、決定論的検証を通し、顧客のアクションゲート経由で宣言を提出して、実体のあるエンティティグラフを育てる。5役のパイプライン（棚卸役→モデル設計役→規律ゲート→宣言提案担当→台帳編集）で、読み取り専用・データ行の書き込みゼロ・v0 は作成のみ。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "ontology-readonly-profiling", "description": "只读取证流程：数据集清单→表结构→逐列四证据（唯一性/空值率/值域样本/外键候选）。" }, { "name": "ontology-draft-deterministic-rules", "description": "草案自洽五判据：主键⊆列快照、关系目标∈主键集合、四元组唯一、1:n 与 n:1 互斥、中间集值域。" }, { "name": "ontology-action-proposal-protocol", "description": "声明提案协议：一件一单、全文件正文、不重试刷屏、未在册动作走 escalate。" } ], "mcpServers": [ { "registryName": "wire-ops" } ] } }, { "id": "Y99VUN6RdWs98ocK0ZAN8A", "slug": "pack-y99vun6rdws98ock0zan8a", "category": "pack", "marketCategory": "fleet", "tags": [ "团队", "识律", "营销" ], "canonical": { "brief": "Lex marketing content cluster template (content-factory variant): collector (customer scenarios and industry pain points) → synthesizer (magnitude wording, no precise numbers) → gatekeeper (red-line review, mandatory) → producer (drafts to the site/CMS; publishing goes through the action-gate propos" }, "human": { "zh-CN": { "name": "识律营销团队", "summary": "lex 营销内容团队模板（content-factory 变体）：采集位（客户场景与行业痛点）→合成位（量级话术，不用精确数字）→闸位（红线审读必设）→产出位（落官网/CMS 草稿，发布走动作闸提案不直发）。装包后经壹座「团队→从模板实例化」建队；纪律来自共用层四包（同 cell 装）：fleet-intel-gather/fleet-brief-report/fleet-data-qa/fleet-board-contribution。" }, "en": { "name": "Lex Marketing", "summary": "Lex marketing content cluster template (content-factory variant): collector (customer scenarios and industry pain points) → synthesizer (magnitude wording, no precise numbers) → gatekeeper (red-line review, mandatory) → producer (drafts to the site/CMS; publishing goes through the action-gate proposal, never direct). Install, then instantiate from the template; discipline comes from the four shared packs (install in the same cell)." }, "es": { "name": "Marketing de Lex", "summary": "Plantilla de clúster de contenido de marketing de Lex (variante content-factory): recolector (escenarios de cliente y problemas del sector) → sintetizador (lenguaje de magnitud, sin cifras precisas) → guardián (revisión de líneas rojas, obligatoria) → productor (borradores para el sitio/CMS; la publicación pasa por la propuesta de la compuerta de acciones, nunca directa). Instala y crea la instancia desde la plantilla; la disciplina viene de los cuatro packs compartidos (en la misma celda)." }, "fr": { "name": "Marketing Lex", "summary": "Modèle de cluster de contenu marketing Lex (variante content-factory) : collecteur (scénarios clients et points de douleur du secteur) → synthétiseur (formulations de magnitude, sans chiffres précis) → gardien (revue des lignes rouges, obligatoire) → producteur (brouillons pour le site/CMS ; la publication passe par la proposition de la porte d'actions, jamais en direct). Installez puis instanciez depuis le modèle ; la discipline vient des quatre packs partagés (même cellule)." }, "ja": { "name": "Lex マーケティングチーム", "summary": "Lex マーケティングコンテンツのクラスタテンプレート（content-factory 変種）：収集役（顧客シナリオと業界の課題）→統合役（桁感の表現、正確な数字は使わない）→ゲート役（レッドライン審査は必須）→制作役（サイト／CMS のドラフトを作成、公開はアクションゲートの提案経由で直接公開しない）。インストール後にテンプレートからインスタンス化。規律は共用4パック（同じセルに導入）から供給。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "legal-marketing-magnitude-wording", "description": "量级话术对照表：各类对外指标的合规模级表述与禁用精确数字清单。" }, { "name": "legal-product-claim-checklist", "description": "识律产品能力表述清单：可写功能事实、禁写效果承诺与案件结论。" }, { "name": "lex-scenario-pitch-card", "description": "选题卡模板：场景/痛点/产品面/成稿理由四栏，一题一卡。" } ], "mcpServers": [ { "registryName": "wire-build-mcp" } ] } }, { "id": "1HbpcL4LaB-csKpUQgN2Ew", "slug": "pack-1hbpcl4lab-cskpuqgn2ew", "category": "pack", "marketCategory": "fleet", "tags": [ "团队", "识律", "政心" ], "canonical": { "brief": "Lex Zhengxin watch cluster template: the monitor patrols the real V-Dem/MARPOR libraries by domain (fd-zhengxin-reference read face), the synthesizer cross-checks the weekly report (major shifts escalate), and the producer lands the atlas draft. Install, then instantiate from the template; disciplin" }, "human": { "zh-CN": { "name": "识律政心监测团队", "summary": "识律政心监测团队模板：监视位分域巡检 V-Dem/MARPOR 真库（fd-zhengxin-reference 读面）→合成位周报交叉核对（重大异动 escalate）→产出位落图集草稿。装包后经壹座「团队→从模板实例化」建队；岗位纪律来自共用层智能体（需同 cell 安装）：fleet-data-qa/fleet-brief-report/fleet-board-contribution。" }, "en": { "name": "Lex Zhengxin Watch", "summary": "Lex Zhengxin watch cluster template: the monitor patrols the real V-Dem/MARPOR libraries by domain (fd-zhengxin-reference read face), the synthesizer cross-checks the weekly report (major shifts escalate), and the producer lands the atlas draft. Install, then instantiate from the template; discipline comes from the shared packs (same cell): fleet-data-qa / fleet-brief-report / fleet-board-contribution." }, "es": { "name": "Monitor de Zhengxin de Lex", "summary": "Plantilla de clúster de monitor de Zhengxin de Lex: el monitor patrulla por dominio las bibliotecas reales V-Dem/MARPOR (superficie de lectura fd-zhengxin-reference), el sintetizador cruza el informe semanal (los cambios mayores escalan) y el productor guarda el borrador del atlas. Instala y crea la instancia desde la plantilla; la disciplina viene de los packs compartidos (misma celda): fleet-data-qa / fleet-brief-report / fleet-board-contribution." }, "fr": { "name": "Veille Zhengxin Lex", "summary": "Modèle de cluster de veille Zhengxin Lex : le moniteur parcourt par domaine les bases réelles V-Dem/MARPOR (surface de lecture fd-zhengxin-reference), le synthétiseur recoupe le rapport hebdomadaire (les changements majeurs escaladent) et le producteur dépose le brouillon de l'atlas. Installez puis instanciez depuis le modèle ; la discipline vient des packs partagés (même cellule) : fleet-data-qa / fleet-brief-report / fleet-board-contribution." }, "ja": { "name": "Lex 政心モニタリングチーム", "summary": "Lex 政心モニタリングクラスタのテンプレート：監視役が V-Dem／MARPOR の実データベースを領域別に巡回（fd-zhengxin-reference の読み取り面）→統合役が週報をクロスチェック（重大な異変はエスカレーション）→制作役が図集の下書きを保存。インストール後にテンプレートからインスタンス化。規律は共用パック（同じセルに導入）：fleet-data-qa / fleet-brief-report / fleet-board-contribution。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "zx-source-readiness-gate", "description": "六源就绪门：source_status 判就绪/延迟/未就绪三档，未就绪整源跳过并如实记录。" }, { "name": "zx-timeseries-anomaly-card", "description": "异动卡五栏：国家/指标/年份对/差值/快照版本，阈值来源必须可复核。" } ], "mcpServers": [ { "registryName": "fd-zhengxin-reference" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "ZCot-b6HsnJRTTDKr_aunQ", "slug": "pack-zcot-b6hsnjrttdkr-aunq", "category": "pack", "marketCategory": "fleet", "tags": [ "团队", "识律", "法规" ], "canonical": { "brief": "Lex statute-update cluster template: the monitor scans new/amended/repealed statutes daily (fd-legal-search-mcp authoritative statute face, 32,296 as the single source of truth), the synthesizer writes the update brief with effective-date impact notes (citation number + article locator, re-checkable" }, "human": { "zh-CN": { "name": "法规追踪团队", "summary": "识律法规更新团队模板：监视位每日扫新颁/修订/废止（fd-legal-search-mcp 权威法规面，32,296 口径单一事实源）→合成位更新简报+生效影响标注（法条引用文号+条款定位可回查）→产出位落简报草稿；无更新日报空。装包后经壹座「团队→从模板实例化」建队；纪律来自共用层智能体（同 cell 装）：fleet-intel-gather/fleet-brief-report/fleet-board-contribution。" }, "en": { "name": "Lex Statute Update", "summary": "Lex statute-update cluster template: the monitor scans new/amended/repealed statutes daily (fd-legal-search-mcp authoritative statute face, 32,296 as the single source of truth), the synthesizer writes the update brief with effective-date impact notes (citation number + article locator, re-checkable), and the producer lands the brief draft; no updates means an empty daily report. Install, then instantiate from the template; discipline comes from the shared packs (same cell): fleet-intel-gather / fleet-brief-report / fleet-board-contribution." }, "es": { "name": "Actualización Normativa de Lex", "summary": "Plantilla de clúster de actualización normativa de Lex: el monitor revisa cada día nuevas normas, modificaciones y derogaciones (superficie normativa autoritativa de fd-legal-search-mcp, 32.296 como única fuente de verdad), el sintetizador redacta el informe de actualización con notas de impacto (número de cita + localización del artículo, verificable) y el productor guarda el borrador; si no hay novedades, el informe diario va vacío. Instala y crea la instancia desde la plantilla; la disciplina viene de los packs compartidos (misma celda): fleet-intel-gather / fleet-brief-report / fleet-board-contribution." }, "fr": { "name": "Veille réglementaire Lex", "summary": "Modèle de cluster de veille réglementaire Lex : le moniteur examine chaque jour les textes nouveaux, modifiés et abrogés (surface normative faisant autorité de fd-legal-search-mcp, 32 296 comme source unique de vérité), le synthétiseur rédige le brief de mise à jour avec les impacts d'entrée en vigueur (numéro de citation + localisation de l'article, vérifiable) et le producteur dépose le brouillon ; sans nouveauté, le rapport quotidien est vide. Installez puis instanciez depuis le modèle ; la discipline vient des packs partagés (même cellule) : fleet-intel-gather / fleet-brief-report / fleet-board-contribution." }, "ja": { "name": "Lex 法令追跡チーム", "summary": "Lex 法令追跡クラスタのテンプレート：監視役が毎日、新規制定・改正・廃止を走査（fd-legal-search-mcp の権威ある法令面、32,296 を単一の事実源とする）→統合役が更新ブリーフと効力影響の注記（法令番号＋条項位置で再確認可能）を作成→制作役がブリーフの下書きを保存。更新がなければ日報は空。インストール後にテンプレートからインスタンス化。規律は共用パック（同じセルに導入）：fleet-intel-gather / fleet-brief-report / fleet-board-contribution。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "statute-scan-status-taxonomy", "description": "新颁/修订/废止三分类机械判据：题名关键词+状态字段，判据可复核。" }, { "name": "statute-reference-resolution", "description": "条级定位解析：law_reference 解析「某法第 N 条」到文号+law_id+条号，失败标不可解。" } ], "mcpServers": [ { "registryName": "fd-legal-search-mcp" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "4z4dMrtNbTn2WcE-45QPfw", "slug": "pack-4z4dmrtnbtn2wce-45qpfw", "category": "pack", "marketCategory": "fleet-shared", "tags": [ "团队", "共用层", "情报", "巡检" ], "canonical": { "brief": "Shared fleet layer: collection discipline for intel patrol posts — fd-info references, source-list patrol, read-the-original before summarizing, dedupe, empty results carry a reason." }, "human": { "zh-CN": { "name": "情报采集", "summary": "团队共用智能体：情报巡检岗的采集纪律——fd-info 检索面全套引用（list_since/get_item/search_info/locate_evidence/stream_health），按源清单巡检、原文先行、去重要点化、空结果带原因、未取得不编造。" }, "en": { "name": "Intel Gathering", "summary": "Shared fleet layer: collection discipline for intel patrol posts — fd-info references, source-list patrol, read-the-original before summarizing, dedupe, empty results carry a reason." }, "es": { "name": "Recolección de Inteligencia", "summary": "Agente compartido del equipo: disciplina de recolección para el rol de patrulla de inteligencia — referencias completas a la superficie de búsqueda fd-info (list_since/get_item/search_info/locate_evidence/stream_health), patrulla por lista de fuentes, leer el original antes de resumir, deduplicar, resultados vacíos con motivo y no inventar lo no obtenido." }, "fr": { "name": "Collecte de renseignement", "summary": "Agent partagé d'équipe : discipline de collecte pour le rôle de patrouille de renseignement — références complètes à la surface de recherche fd-info (list_since/get_item/search_info/locate_evidence/stream_health), patrouille selon la liste de sources, lire l'original avant de résumer, dédupliquer, résultats vides avec motif et ne jamais inventer l'absent." }, "ja": { "name": "情報収集", "summary": "チーム共用のエージェント：情報巡回役の収集規律——fd-info 検索面一式を参照（list_since／get_item／search_info／locate_evidence／stream_health）、ソース一覧に沿って巡回、原文先行、重複排除と要点化、空結果には理由、取得不能を捏造しない。" } }, "whenToUse": "Browse under \"fleet-shared\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "fleet-intel-gather", "description": "情报采集纪律：按源清单巡检、原文先行、去重要点化、空结果带原因分类。" } ], "mcpServers": [ { "registryName": "fd-info" } ] } }, { "id": "Mo3_aWe7tF2eFVzkqFT9mA", "slug": "pack-mo3-awe7tf2efvzkqft9ma", "category": "pack", "marketCategory": "fleet-shared", "tags": [ "团队", "共用层", "看板", "投稿" ], "canonical": { "brief": "Shared fleet layer: artifact-write discipline for contribution posts — wire-build references, draft-first, source_ref triplets, column pre-check, four-way failure attribution." }, "human": { "zh-CN": { "name": "看板投稿", "summary": "团队共用智能体：投稿岗的产物写面纪律——wire-build 全套引用（建集/表单/记录视图/看板/文档/发布/治理），草稿优先、source_ref 三元组、列名预检、失败四分类。装配于任何板块的团队投稿岗。" }, "en": { "name": "Board Contribution", "summary": "Shared fleet layer: artifact-write discipline for contribution posts — wire-build references, draft-first, source_ref triplets, column pre-check, four-way failure attribution." }, "es": { "name": "Contribución al Tablero", "summary": "Agente compartido: disciplina de escritura de artefactos para el rol de contribución — referencias a wire-build (conjuntos, formularios, vistas, tableros, documentos, publicación y gobernanza), borrador primero, tripleta source_ref, verificación de columnas y cuatro categorías de fallo. Para cualquier dominio." }, "fr": { "name": "Contribution au tableau de bord", "summary": "Agent partagé : discipline d'écriture d'artefacts pour le rôle de contribution — références à wire-build (jeux de données, formulaires, vues, tableaux de bord, documents, publication et gouvernance), brouillon d'abord, triplet source_ref, contrôle des colonnes et quatre catégories d'échec. Pour tout domaine." }, "ja": { "name": "ボード投稿", "summary": "チーム共用のエージェント：投稿役の成果物書き込み規律——wire-build 一式を参照（データセット作成／フォーム／レコードビュー／ボード／ドキュメント／公開／ガバナンス）、下書き優先、source_ref 三元組、列名の事前検査、失敗の四分類。どの領域のチームにも投稿役として組み込める。" } }, "whenToUse": "Browse under \"fleet-shared\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "fleet-board-contribution", "description": "看板投稿纪律：草稿优先、source_ref 三元组、列名预检、失败四分类、零半写处理。" } ], "mcpServers": [ { "registryName": "wire-build-mcp" } ] } }, { "id": "O30LA92kq1D5uuNIBGKsZQ", "slug": "pack-o30la92kq1d5uunibgkszq", "category": "facade", "marketCategory": "fleet", "tags": [ "团队", "构造师", "蓝图" ], "canonical": { "brief": "Blueprint construction face: compiles a natural-language vertical brief into a blueprint draft (boundary/sources/fleet/boards/monitoring/acceptance), dry-runs the existing validators, shows the diff and waits for the owner's approval — only then mints and instantiates. No new platform capability." }, "human": { "zh-CN": { "name": "蓝图构造师", "summary": "蓝图构造间构造面：把自然语言垂直描述编译成蓝图草稿（板块边界/数据源/团队编制/看板骨架/监控/验收口径），复用既有校验器 dry-run，diff 呈现后停等主人批准——批准才铸造上架与实例化。只引用 fleet-architect 与 facet-market 既有工具面，蓝图格式以平台校验器为准。" }, "en": { "name": "Vertical Constructor", "summary": "Blueprint construction face: compiles a natural-language vertical brief into a blueprint draft (boundary/sources/fleet/boards/monitoring/acceptance), dry-runs the existing validators, shows the diff and waits for the owner's approval — only then mints and instantiates. No new platform capability." }, "es": { "name": "Constructor Vertical", "summary": "Superficie de construcción de planos: compila una descripción vertical en lenguaje natural en un borrador de plano (límites del dominio, fuentes de datos, composición del equipo, esqueleto del tablero, monitoreo y criterios de aceptación), ejecuta en seco los validadores existentes, muestra el diff y espera la aprobación del propietario — solo entonces acuña e instancia. No añade capacidades nuevas a la plataforma." }, "fr": { "name": "Constructeur vertical", "summary": "Surface de construction de plans : compile une description verticale en langage naturel en un brouillon de plan (limites du domaine, sources de données, composition de l'équipe, squelette du tableau de bord, surveillance et critères d'acceptation), exécute les validateurs existants à blanc, affiche le diff et attend la validation du propriétaire — ensuite seulement, il forge et instancie. N'ajoute aucune capacité à la plateforme." }, "ja": { "name": "垂直コンストラクタ", "summary": "ブループリント構築面：自然言語の垂直記述をブループリント草案（領域境界／データソース／チーム編成／ボード骨子／監視／受け入れ基準）にコンパイルし、既存の検証器で dry-run、diff を提示して主人の承認を待つ——承認後にのみ鋳造・公開・インスタンス化する。新たなプラットフォーム機能は追加しない。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "harness-constructor", "description": "垂直蓝图编制纪律：NL→草稿→既有校验器→diff→停等批准→批准后铸造；字段契约与八舰同源。" } ], "mcpServers": [ { "registryName": "fleet-architect" }, { "registryName": "facet-market" } ] } }, { "id": "_Rl8i99CS5nwuARVbp0D5Q", "slug": "pack-rl8i99cs5nwuarvbp0d5q", "category": "pack", "marketCategory": "fleet-shared", "tags": [ "团队", "共用层", "评测", "质量" ], "canonical": { "brief": "Shared fleet layer: quality-gate discipline — gold-set comparison (exact/direction/coverage), KPI reporting, drift escalation, eval reports landed as drafts." }, "human": { "zh-CN": { "name": "评测哨", "summary": "团队共用智能体：评测岗的质量对拍纪律——按 gold 清单对拍目标岗产物（exact/方向/覆盖率），KPI 报数（hit_rate/direction_rate/coverage），漂移超阈值 escalate，评测报告落草稿。可选挂载岗。" }, "en": { "name": "Eval Sentinel", "summary": "Shared fleet layer: quality-gate discipline — gold-set comparison (exact/direction/coverage), KPI reporting, drift escalation, eval reports landed as drafts." }, "es": { "name": "Centinela de Evaluación", "summary": "Agente compartido del equipo: disciplina de control de calidad para el rol de evaluación — comparación con la lista gold (exacto/dirección/cobertura), reporte de KPI (hit_rate/direction_rate/coverage), escalado de deriva sobre el umbral y guardado del informe de evaluación como borrador. Puesto opcional." }, "fr": { "name": "Sentinelle d'évaluation", "summary": "Agent partagé d'équipe : discipline de contrôle qualité pour le rôle d'évaluation — comparaison au jeu gold (exact/direction/couverture), reporting KPI (hit_rate/direction_rate/coverage), escalade des dérives au-delà du seuil et dépôt du rapport d'évaluation en brouillon. Poste optionnel." }, "ja": { "name": "評価番", "summary": "チーム共用のエージェント：評価役の品質対照規律——gold リストに沿って対象役の成果物を対照（exact／方向／カバレッジ）、KPI 報告（hit_rate／direction_rate／coverage）、閾値超のドリフトは escalate、評価レポートは下書きで保存。任意で装着する役割。" } }, "whenToUse": "Browse under \"fleet-shared\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "fleet-eval-sentinel", "description": "评测哨纪律：gold 对拍、KPI 报数、漂移预警 escalate、评测报告落草稿。" } ], "mcpServers": [ { "registryName": "wire-ops" }, { "registryName": "wire-build-mcp" } ] } }, { "id": "y2_QsuKxV8dZFr-39_0zRQ", "slug": "pack-y2-qsukxv8dzfr-39-0zrq", "category": "official-data", "marketCategory": "fleet-shared", "tags": [ "团队", "共用层", "数据", "问答" ], "canonical": { "brief": "Shared fleet layer: read-only analysis discipline — dataset read faces, every number traceable, scope stated, honest empty results, chart-contract v2 selection." }, "human": { "zh-CN": { "name": "数据问答", "summary": "团队共用智能体：数据问答/分析岗的只读纪律——数据集读面引用（列表/查询/聚合/行读），数字可溯源、口径注明、空结果如实、图表契约 v2 选型。只读，不含写面。" }, "en": { "name": "Data Q&A", "summary": "Shared fleet layer: read-only analysis discipline — dataset read faces, every number traceable, scope stated, honest empty results, chart-contract v2 selection." }, "es": { "name": "Preguntas y Respuestas de Datos", "summary": "Agente compartido: solo lectura para Q&A y análisis — superficies de lectura (lista, consulta, agregación, filas), cifras trazables, criterio declarado, vacíos honestos y contrato de gráficos v2. Sin escritura." }, "fr": { "name": "Questions-réponses sur les données", "summary": "Agent partagé : lecture seule pour Q&R et analyse — surfaces de lecture (liste, requête, agrégation, lignes), chiffres traçables, critère explicite, vides honnêtes et contrat de graphiques v2. Sans écriture." }, "ja": { "name": "データQ&A", "summary": "チーム共用のエージェント：データQ&A／分析役の読み取り専用規律——データセット読み取り面の参照（一覧／クエリ／集計／行読み）、数字は追跡可能、口径を明記、空結果は正直に、チャート契約 v2 の選型。読み取り専用で書き込み面を含まない。" } }, "whenToUse": "Browse under \"fleet-shared\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "fleet-data-qa", "description": "数据问答纪律：先探后答、数字可溯源、口径注明、图表契约选型、只读。" } ], "mcpServers": [ { "registryName": "wire-ops" } ] } }, { "id": "F8gbhNaEG6c2rDz468MwaA", "slug": "pack-f8gbhnaeg6c2rdz468mwaa", "category": "pack", "marketCategory": "fleet-shared", "tags": [ "团队", "共用层", "简报", "报数" ], "canonical": { "brief": "Shared fleet layer: reporting and briefing discipline for brief/chief posts — report_status slots aligned to the fleet KPIs, morning/weekly brief skeletons, escalate norms, and never fabricate what you could not obtain." }, "human": { "zh-CN": { "name": "简报与报数", "summary": "团队共用智能体：合成位/主编岗的报数与简报纪律——report_status 槽位报数（与团队 KPI 槽位逐字对齐）、晨报/周报骨架、escalate 拍板规范、未取得不编造。装配于任何板块的团队合成位。" }, "en": { "name": "Briefing & Reporting", "summary": "Shared fleet layer: reporting and briefing discipline for brief/chief posts — report_status slots aligned to the fleet KPIs, morning/weekly brief skeletons, escalate norms, and never fabricate what you could not obtain." }, "es": { "name": "Informes y Reporte", "summary": "Agente compartido: disciplina de informe y reporte para los roles de síntesis y edición — reporte por slots con report_status (claves idénticas a los KPI del equipo), esqueleto de informe diario/semanal, normas de escalate y nunca inventar lo no obtenido. Para cualquier dominio." }, "fr": { "name": "Brief et reporting", "summary": "Agent partagé : discipline de brief et de reporting pour les rôles de synthèse et de rédaction en chef — reporting par slots avec report_status (clés identiques aux KPI de l'équipe), squelette de brief quotidien/hebdomadaire, normes d'escalade et ne jamais inventer l'absent. Pour tout domaine." }, "ja": { "name": "ブリーフと報告", "summary": "チーム共用のエージェント：統合役／編集長役の報告とブリーフ規律——report_status のスロット報告（チーム KPI スロットと逐語一致）、朝報／週報の骨子、escalate の承認規範、取得できなかったことを捏造しない。どの領域のチームの統合役にも組み込める。" } }, "whenToUse": "Browse under \"fleet-shared\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "fleet-brief-report", "description": "舰队简报与报数纪律：槽位报数、晨报/周报骨架、escalate 规范、未取得不编造。" } ], "mcpServers": [] } }, { "id": "v8rZ8F7E92MrssZ92ygnXQ", "slug": "pack-v8rz8f7e92mrssz92ygnxq", "category": "pack", "marketCategory": "fleet", "tags": [], "canonical": { "brief": "External-facing product-introduction team: one front-desk “Product Guide” plus five product-line specialists (FindData·Base / Lex / Wire / Facet / Constellation). Orchestration is a dispatch star from the guide to each specialist (a conversational team with no cron cadence — once installed it simply" }, "human": { "zh-CN": { "name": "壹座产品介绍", "summary": "对外产品介绍团队：前台「产品向导」+ 五条产品线专家（寻数·壹座/识律/柏讯/谦面/萬星），编导为前台→各专家的 dispatch 星型（会话型团队，无 cron 节奏——装上即等访客提问）。知识来源：www.finddatatech.cloud 官网中文页 /zh/（原始抓取，抓取日期 2026-10-30）与本仓公开 README（github.com/FindDataTechnology/platform）；官网改版时人工同步本种子并重跑守恒与红线测试。结构注：六岗全部自载（maxAgents 50，2026-10-30 用户裁决放宽，75df6914）——包自带完整团队，外来安装者零预创建、无降级路径。KPI 槽位：会话型团队无运营语义，不设（kpiSlots: []）。预校验兼容性结论（2026-10-30 实证）：无 cadence/显式 cadence:null 的岗位直接通过 parseFleetTemplate——design D2 的值守 cron 回退方案无需启用。本包声明 publicFacing=true（对外团队）：其岗位运行时只挂公开工具面（fleet 桥 + 市场公开面），禁用工作区文件工具与名册枚举工具——内部车道名不得进入访客可见的回答。红线：全部 persona 只含公开信息，词表 v4（智能体团队/小组/智能体/智能体市场），禁内部运营信息——scripts/test-product-intro-seed.mjs 以文本断言钉住。五线名以官网中文页原文为准（寻数·壹座/识律/柏讯/谦面/萬星）。" }, "en": { "name": "FindData Product Introduction", "summary": "External-facing product-introduction team: one front-desk “Product Guide” plus five product-line specialists (FindData·Base / Lex / Wire / Facet / Constellation). Orchestration is a dispatch star from the guide to each specialist (a conversational team with no cron cadence — once installed it simply waits for visitor questions). Knowledge source: the official Chinese site www.finddatatech.cloud /zh/ (fetched verbatim on 2026-10-30) and the public README (github.com/FindDataTechnology/platform); when the site changes, sync this seed by hand and re-run the conservation and red-line tests. Structure note: all six posts are self-carried (maxAgents 50) — the pack ships the complete team, so an outside installer needs zero pre-creation and has no degraded path. KPI slots: a conversational team carries no operations semantics, so none are declared (kpiSlots: []). This pack declares publicFacing=true: at runtime its posts mount only the public tool surface (fleet bridge + public market), with workspace file tools and roster-enumeration tools disabled — internal lane names must never reach a visitor-visible answer. Red line: every persona contains public information only, uses vocabulary v4 (agent team / squad / agent / agent market), and never internal operations detail — pinned by text assertions in scripts/test-product-intro-seed.mjs. Product-line names follow the official Chinese site verbatim." }, "es": { "name": "壹座产品介绍", "summary": "对外产品介绍团队：前台「产品向导」+ 五条产品线专家（寻数·壹座/识律/柏讯/谦面/萬星），编导为前台→各专家的 dispatch 星型（会话型团队，无 cron 节奏——装上即等访客提问）。知识来源：www.finddatatech.cloud 官网中文页 /zh/（原始抓取，抓取日期 2026-10-30）与本仓公开 README（github.com/FindDataTechnology/platform）；官网改版时人工同步本种子并重跑守恒与红线测试。结构注：六岗全部自载（maxAgents 50，2026-10-30 用户裁决放宽，75df6914）——包自带完整团队，外来安装者零预创建、无降级路径。KPI 槽位：会话型团队无运营语义，不设（kpiSlots: []）。预校验兼容性结论（2026-10-30 实证）：无 cadence/显式 cadence:null 的岗位直接通过 parseFleetTemplate——design D2 的值守 cron 回退方案无需启用。本包声明 publicFacing=true（对外团队）：其岗位运行时只挂公开工具面（fleet 桥 + 市场公开面），禁用工作区文件工具与名册枚举工具——内部车道名不得进入访客可见的回答。红线：全部 persona 只含公开信息，词表 v4（智能体团队/小组/智能体/智能体市场），禁内部运营信息——scripts/test-product-intro-seed.mjs 以文本断言钉住。五线名以官网中文页原文为准（寻数·壹座/识律/柏讯/谦面/萬星）。" }, "fr": { "name": "壹座产品介绍", "summary": "对外产品介绍团队：前台「产品向导」+ 五条产品线专家（寻数·壹座/识律/柏讯/谦面/萬星），编导为前台→各专家的 dispatch 星型（会话型团队，无 cron 节奏——装上即等访客提问）。知识来源：www.finddatatech.cloud 官网中文页 /zh/（原始抓取，抓取日期 2026-10-30）与本仓公开 README（github.com/FindDataTechnology/platform）；官网改版时人工同步本种子并重跑守恒与红线测试。结构注：六岗全部自载（maxAgents 50，2026-10-30 用户裁决放宽，75df6914）——包自带完整团队，外来安装者零预创建、无降级路径。KPI 槽位：会话型团队无运营语义，不设（kpiSlots: []）。预校验兼容性结论（2026-10-30 实证）：无 cadence/显式 cadence:null 的岗位直接通过 parseFleetTemplate——design D2 的值守 cron 回退方案无需启用。本包声明 publicFacing=true（对外团队）：其岗位运行时只挂公开工具面（fleet 桥 + 市场公开面），禁用工作区文件工具与名册枚举工具——内部车道名不得进入访客可见的回答。红线：全部 persona 只含公开信息，词表 v4（智能体团队/小组/智能体/智能体市场），禁内部运营信息——scripts/test-product-intro-seed.mjs 以文本断言钉住。五线名以官网中文页原文为准（寻数·壹座/识律/柏讯/谦面/萬星）。" }, "ja": { "name": "壹座产品介绍", "summary": "对外产品介绍团队：前台「产品向导」+ 五条产品线专家（寻数·壹座/识律/柏讯/谦面/萬星），编导为前台→各专家的 dispatch 星型（会话型团队，无 cron 节奏——装上即等访客提问）。知识来源：www.finddatatech.cloud 官网中文页 /zh/（原始抓取，抓取日期 2026-10-30）与本仓公开 README（github.com/FindDataTechnology/platform）；官网改版时人工同步本种子并重跑守恒与红线测试。结构注：六岗全部自载（maxAgents 50，2026-10-30 用户裁决放宽，75df6914）——包自带完整团队，外来安装者零预创建、无降级路径。KPI 槽位：会话型团队无运营语义，不设（kpiSlots: []）。预校验兼容性结论（2026-10-30 实证）：无 cadence/显式 cadence:null 的岗位直接通过 parseFleetTemplate——design D2 的值守 cron 回退方案无需启用。本包声明 publicFacing=true（对外团队）：其岗位运行时只挂公开工具面（fleet 桥 + 市场公开面），禁用工作区文件工具与名册枚举工具——内部车道名不得进入访客可见的回答。红线：全部 persona 只含公开信息，词表 v4（智能体团队/小组/智能体/智能体市场），禁内部运营信息——scripts/test-product-intro-seed.mjs 以文本断言钉住。五线名以官网中文页原文为准（寻数·壹座/识律/柏讯/谦面/萬星）。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "public-info-discipline", "description": "对外介绍公开信息纪律：只官网与公开仓库口径，内部名零出现，不确定如实说。" }, { "name": "intro-triage-routing", "description": "前台分诊路由卡：话题到专家的映射与转介话术。" } ], "mcpServers": [] } }, { "id": "NoBBXkvxZnJNHlCrAwWp-w", "slug": "pack-nobbxkvxznjnhlcrawwp-w", "category": "facade", "marketCategory": "fleet", "tags": [], "canonical": { "brief": "壹座聊天内的团队自然语言编排面：把用户的话编译成团队结构草稿，经校验与成本预估呈 diff，用户批准后落地。管编制不管作战——不做运行时指挥。铸造上架时 packId/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 pack 清单校验。" }, "human": { "zh-CN": { "name": "团队架构师", "summary": "壹座聊天内的团队自然语言编排面：把用户的话编译成团队结构草稿，经校验与成本预估呈 diff，用户批准后落地。管编制不管作战——不做运行时指挥。铸造上架时 packId/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 pack 清单校验。" }, "en": { "name": "团队架构师", "summary": "壹座聊天内的团队自然语言编排面：把用户的话编译成团队结构草稿，经校验与成本预估呈 diff，用户批准后落地。管编制不管作战——不做运行时指挥。铸造上架时 packId/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 pack 清单校验。" }, "es": { "name": "团队架构师", "summary": "壹座聊天内的团队自然语言编排面：把用户的话编译成团队结构草稿，经校验与成本预估呈 diff，用户批准后落地。管编制不管作战——不做运行时指挥。铸造上架时 packId/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 pack 清单校验。" }, "fr": { "name": "团队架构师", "summary": "壹座聊天内的团队自然语言编排面：把用户的话编译成团队结构草稿，经校验与成本预估呈 diff，用户批准后落地。管编制不管作战——不做运行时指挥。铸造上架时 packId/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 pack 清单校验。" }, "ja": { "name": "团队架构师", "summary": "壹座聊天内的团队自然语言编排面：把用户的话编译成团队结构草稿，经校验与成本预估呈 diff，用户批准后落地。管编制不管作战——不做运行时指挥。铸造上架时 packId/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 pack 清单校验。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "fleet-authoring-flow", "description": "团队编排流程卡：需求→草稿→diff 呈现→批准→落地。" }, { "name": "fleet-edge-vocabulary", "description": "编边词汇卡：主管式/流水线/单边派发/草稿引用的编译对照。" } ], "mcpServers": [] } }, { "id": "5y0_Gislp_OgElqq4zPswA", "slug": "pack-5y0-gislp-ogelqq4zpswa", "category": "facade", "marketCategory": "text", "tags": [ "Productivity" ], "canonical": { "brief": "Edit PDFs with natural-language instructions using the nano-pdf CLI." }, "human": { "zh-CN": { "name": "nano-pdf", "summary": "Edit PDFs with natural-language instructions using the nano-pdf CLI." }, "en": { "name": "nano-pdf", "summary": "Edit PDFs with natural-language instructions using the nano-pdf CLI." }, "es": { "name": "nano-pdf", "summary": "Edit PDFs with natural-language instructions using the nano-pdf CLI." }, "fr": { "name": "nano-pdf", "summary": "Edit PDFs with natural-language instructions using the nano-pdf CLI." }, "ja": { "name": "nano-pdf", "summary": "Edit PDFs with natural-language instructions using the nano-pdf CLI." } }, "whenToUse": "Browse under \"text\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "nano-pdf", "description": "Edit PDFs with natural-language instructions using the nano-pdf CLI." } ], "mcpServers": [] } }, { "id": "egaYyOaZVTf0FWlLvaGAQA", "slug": "pack-egayyoazvtf0fwllvagaqa", "category": "pack", "marketCategory": "research", "tags": [ "柏讯", "投研", "团队" ], "canonical": { "brief": "三员投研团队：宏观分析（环境）、估值分析（价值）、投委会多空质询（博弈）三个独立镜头审读同一问题。团队任务下扇出聚合；单聊可切换任一角色。舆情/风控等更多角色可入动态团队叠加。" }, "human": { "zh-CN": { "name": "柏讯 · 投研三镜头", "summary": "三员投研团队：宏观分析（环境）、估值分析（价值）、投委会多空质询（博弈）三个独立镜头审读同一问题。团队任务下扇出聚合；单聊可切换任一角色。舆情/风控等更多角色可入动态团队叠加。" }, "en": { "name": "柏讯 · 投研三镜头", "summary": "三员投研团队：宏观分析（环境）、估值分析（价值）、投委会多空质询（博弈）三个独立镜头审读同一问题。团队任务下扇出聚合；单聊可切换任一角色。舆情/风控等更多角色可入动态团队叠加。" }, "es": { "name": "柏讯 · 投研三镜头", "summary": "三员投研团队：宏观分析（环境）、估值分析（价值）、投委会多空质询（博弈）三个独立镜头审读同一问题。团队任务下扇出聚合；单聊可切换任一角色。舆情/风控等更多角色可入动态团队叠加。" }, "fr": { "name": "柏讯 · 投研三镜头", "summary": "三员投研团队：宏观分析（环境）、估值分析（价值）、投委会多空质询（博弈）三个独立镜头审读同一问题。团队任务下扇出聚合；单聊可切换任一角色。舆情/风控等更多角色可入动态团队叠加。" }, "ja": { "name": "柏讯 · 投研三镜头", "summary": "三员投研团队：宏观分析（环境）、估值分析（价值）、投委会多空质询（博弈）三个独立镜头审读同一问题。团队任务下扇出聚合；单聊可切换任一角色。舆情/风控等更多角色可入动态团队叠加。" } }, "whenToUse": "Browse under \"research\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "macro-cycle-analysis", "description": "宏观周期与景气分析。当用户要求判断当前经济周期位置（复苏/过热/滞胀/衰退）、做六域景气扫描（增长/通胀/货币信用/财政/外贸/地产消费）、做领先-同步-滞后指标分层、识别景气拐点、梳理宏观变量到其业务问题的传导链，或做省际补充与国际对比时使用。走柏讯服务端宏观数据，不含收益率曲线与互换分析。" }, { "name": "statistical-inference", "description": "统计检验与探索性分析。当用户要求做描述统计与分布诊断、选假设检验（t/ANOVA/卡方/非参）、判断两组或城市间差异是否显著、做相关性与配对分析、诊断时序平稳性与单位根、做多重比较校正，或对已有的本地 CSV/表格做探索性分析时使用。走柏讯服务端序列取数，价格类序列走本地 DAAS。" }, { "name": "equity-valuation", "description": "股票估值（DCF + 可比公司 + 三表联动）。当用户要求给某家 A 股/港股公司做内在价值或相对估值、算 WACC 与终值、跑敏感性与情景表、给目标价区间、做同业倍数对照与反推市场隐含预期时使用。财务数据走柏讯服务端通道 A，价格与市值走本地 DAAS 通道 B；通道 B 未连接时中止估值，不给目标价。" }, { "name": "earnings-cycle-analysis", "description": "财报周期分析（前瞻 + 复盘）。当用户要求对一家公司做财报前瞻（本次披露类型与时间、该盯哪些科目、超预期/低于预期可能从哪来）、做财报复盘（实际 vs 上年同期、差异由哪个科目驱动、盈利质量如何、对后续季度意味着什么），或梳理连续数期经营动能变化时使用。基于柏讯财报与公告数据，不做技术分析。触发词：财报前瞻、业绩预告、财报复盘、业绩点评、超预期、低于预期、盈利质量、某某公司季报/半年报/年报分析。" }, { "name": "ic-adjudication", "description": "柏讯投委会 · 裁决（入口技能）。当用户要求\"开投委会\"\"对某只股票做一次多空辩论并给结论\"\"裁决多头空头的分歧\"\"把辩论结论落账\"时使用。本技能是套件入口：调度 ic-bull-case 与 ic-bear-case 双方立论，然后按四条裁决纪律给出 Rating 契约实例（冲突不给 hold 须指强弱、证据不足给 review 不降级、裁决不引用任何数字、change_my_mind 必填），并完成落账（账本不可用时明示\"未落账\"）。单 agent 分角色形态，不构成独立视角。" }, { "name": "ic-bull-case", "description": "柏讯投委会 · 多头立论。当用户要求对某只股票构建看多论证、写多头观点、准备投委会辩论的多方意见、把证据集收敛成\"为什么该买\"的最强论证，或作为 ic-adjudication 的立论阶段被调度时使用。本技能不取数、不计算、不裁决——数字只能引用证据集的 evidence_id，论证组织是唯一职责。与 ic-bear-case 对称，空头未发言时不得替对手立论。" }, { "name": "ic-bear-case", "description": "柏讯投委会 · 空头立论（辩论三角色之一，与 ic-bull-case 完全对称）。当用户在投委会辩论流程中需要空头方论证、问「有什么理由不买」「多头论证哪里站不住」「估值凭什么下修」，或作为 ic-adjudication 的立论阶段被调度时使用。本技能不取数、不计算、不裁决——数字只能引用证据集的 evidence_id；不同意多头时只能提出新证据或指出其适用边界，不得改写数字。空头论证为空是合法产出，不得为对称而编造。" } ], "mcpServers": [ { "registryName": "fd-cn-report" }, { "registryName": "fd-daas-mcp" }, { "registryName": "fd-quant" }, { "registryName": "fd-find-data-business-mcp" }, { "registryName": "fd-open-data-mcp" } ] } }, { "id": "_vcfZzuxPgkiRsj5poZIgg", "slug": "pack-vcfzzuxpgkirsj5pozigg", "category": "facade", "marketCategory": "company", "tags": [ "分身", "创作工具", "引导" ], "canonical": { "brief": "和工坊助手聊几句，它把你的知识变成一个可发布的 AI 分身：采访你、整理语料、生成接待人设与 manifest.json，给出发布三步指引。造好的分身部署后，任何人从目录点「和这个分身聊」就能了解你——费用你出（hosted），访客免费。" }, "human": { "zh-CN": { "name": "分身工坊", "summary": "和工坊助手聊几句，它把你的知识变成一个可发布的 AI 分身：采访你、整理语料、生成接待人设与 manifest.json，给出发布三步指引。造好的分身部署后，任何人从目录点「和这个分身聊」就能了解你——费用你出（hosted），访客免费。" }, "en": { "name": "分身工坊", "summary": "和工坊助手聊几句，它把你的知识变成一个可发布的 AI 分身：采访你、整理语料、生成接待人设与 manifest.json，给出发布三步指引。造好的分身部署后，任何人从目录点「和这个分身聊」就能了解你——费用你出（hosted），访客免费。" }, "es": { "name": "分身工坊", "summary": "和工坊助手聊几句，它把你的知识变成一个可发布的 AI 分身：采访你、整理语料、生成接待人设与 manifest.json，给出发布三步指引。造好的分身部署后，任何人从目录点「和这个分身聊」就能了解你——费用你出（hosted），访客免费。" }, "fr": { "name": "分身工坊", "summary": "和工坊助手聊几句，它把你的知识变成一个可发布的 AI 分身：采访你、整理语料、生成接待人设与 manifest.json，给出发布三步指引。造好的分身部署后，任何人从目录点「和这个分身聊」就能了解你——费用你出（hosted），访客免费。" }, "ja": { "name": "分身工坊", "summary": "和工坊助手聊几句，它把你的知识变成一个可发布的 AI 分身：采访你、整理语料、生成接待人设与 manifest.json，给出发布三步指引。造好的分身部署后，任何人从目录点「和这个分身聊」就能了解你——费用你出（hosted），访客免费。" } }, "whenToUse": "Browse under \"company\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "facade-builder-guide", "description": "分身采访与生成流程：四步采访、素材整理纪律、manifest 组装规则、发布指引" }, { "name": "facade-manifest-template", "description": "分身 manifest 骨架：可直填的完整结构，含分身配置块与语料/接待的最小示例" } ], "mcpServers": [] } }, { "id": "56opYgliV9FYAmhmsDIJOw", "slug": "pack-56opygliv9fyamhmsdijow", "category": "pack", "marketCategory": "fleet", "tags": [], "canonical": { "brief": "官方团队模板：每日晨报情报站——科技与市场情报员每日清晨分头巡检，情报主编 8 点合成晨报；另含团队引用岗（existing-cluster，每周一做深度聚合，演示团队 fan-out 引用与派生报数）。演示：双情报源并行、晨报流、cluster 引用岗（公司种子只演示了 a2a）；编导段为 trigger+handoff 链（两情报员巡检完成接棒主编合成）与团队 report 上行。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "human": { "zh-CN": { "name": "晨报情报站", "summary": "官方团队模板：每日晨报情报站——科技与市场情报员每日清晨分头巡检，情报主编 8 点合成晨报；另含团队引用岗（existing-cluster，每周一做深度聚合，演示团队 fan-out 引用与派生报数）。演示：双情报源并行、晨报流、cluster 引用岗（公司种子只演示了 a2a）；编导段为 trigger+handoff 链（两情报员巡检完成接棒主编合成）与团队 report 上行。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "en": { "name": "晨报情报站", "summary": "官方团队模板：每日晨报情报站——科技与市场情报员每日清晨分头巡检，情报主编 8 点合成晨报；另含团队引用岗（existing-cluster，每周一做深度聚合，演示团队 fan-out 引用与派生报数）。演示：双情报源并行、晨报流、cluster 引用岗（公司种子只演示了 a2a）；编导段为 trigger+handoff 链（两情报员巡检完成接棒主编合成）与团队 report 上行。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "es": { "name": "晨报情报站", "summary": "官方团队模板：每日晨报情报站——科技与市场情报员每日清晨分头巡检，情报主编 8 点合成晨报；另含团队引用岗（existing-cluster，每周一做深度聚合，演示团队 fan-out 引用与派生报数）。演示：双情报源并行、晨报流、cluster 引用岗（公司种子只演示了 a2a）；编导段为 trigger+handoff 链（两情报员巡检完成接棒主编合成）与团队 report 上行。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "fr": { "name": "晨报情报站", "summary": "官方团队模板：每日晨报情报站——科技与市场情报员每日清晨分头巡检，情报主编 8 点合成晨报；另含团队引用岗（existing-cluster，每周一做深度聚合，演示团队 fan-out 引用与派生报数）。演示：双情报源并行、晨报流、cluster 引用岗（公司种子只演示了 a2a）；编导段为 trigger+handoff 链（两情报员巡检完成接棒主编合成）与团队 report 上行。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "ja": { "name": "晨报情报站", "summary": "官方团队模板：每日晨报情报站——科技与市场情报员每日清晨分头巡检，情报主编 8 点合成晨报；另含团队引用岗（existing-cluster，每周一做深度聚合，演示团队 fan-out 引用与派生报数）。演示：双情报源并行、晨报流、cluster 引用岗（公司种子只演示了 a2a）；编导段为 trigger+handoff 链（两情报员巡检完成接棒主编合成）与团队 report 上行。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "intel-item-five-elements", "description": "情报员采集与初筛的标准格式：事实、信源、影响面、置信度、建议动作。" }, { "name": "brief-synthesis-template", "description": "主编合成晨报的标准结构：头条、要点、风险预警、信源覆盖度。" } ], "mcpServers": [ { "registryName": "fd-info" } ] } }, { "id": "tkHe8IHE0BuDYYIg5LglUw", "slug": "pack-tkhe8ihe0budyyig5lgluw", "category": "pack", "marketCategory": "fleet", "tags": [], "canonical": { "brief": "官方团队模板：周流水线内容工厂——主编周一主持选题会，选题策划立项、撰稿人周二至四成稿，审校为被动岗（受委托唤醒，演示成员间横向协作的唯一通道 delegation）。演示：周节奏流水线、被动岗受委托唤醒、文档工具面；编导段为 trigger(周一)→handoff 流水线（主编→策划→撰稿）+ 撰稿 dispatch 委托审校、审校 report 上行主编。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "human": { "zh-CN": { "name": "内容流水线", "summary": "官方团队模板：周流水线内容工厂——主编周一主持选题会，选题策划立项、撰稿人周二至四成稿，审校为被动岗（受委托唤醒，演示成员间横向协作的唯一通道 delegation）。演示：周节奏流水线、被动岗受委托唤醒、文档工具面；编导段为 trigger(周一)→handoff 流水线（主编→策划→撰稿）+ 撰稿 dispatch 委托审校、审校 report 上行主编。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "en": { "name": "内容流水线", "summary": "官方团队模板：周流水线内容工厂——主编周一主持选题会，选题策划立项、撰稿人周二至四成稿，审校为被动岗（受委托唤醒，演示成员间横向协作的唯一通道 delegation）。演示：周节奏流水线、被动岗受委托唤醒、文档工具面；编导段为 trigger(周一)→handoff 流水线（主编→策划→撰稿）+ 撰稿 dispatch 委托审校、审校 report 上行主编。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "es": { "name": "内容流水线", "summary": "官方团队模板：周流水线内容工厂——主编周一主持选题会，选题策划立项、撰稿人周二至四成稿，审校为被动岗（受委托唤醒，演示成员间横向协作的唯一通道 delegation）。演示：周节奏流水线、被动岗受委托唤醒、文档工具面；编导段为 trigger(周一)→handoff 流水线（主编→策划→撰稿）+ 撰稿 dispatch 委托审校、审校 report 上行主编。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "fr": { "name": "内容流水线", "summary": "官方团队模板：周流水线内容工厂——主编周一主持选题会，选题策划立项、撰稿人周二至四成稿，审校为被动岗（受委托唤醒，演示成员间横向协作的唯一通道 delegation）。演示：周节奏流水线、被动岗受委托唤醒、文档工具面；编导段为 trigger(周一)→handoff 流水线（主编→策划→撰稿）+ 撰稿 dispatch 委托审校、审校 report 上行主编。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "ja": { "name": "内容流水线", "summary": "官方团队模板：周流水线内容工厂——主编周一主持选题会，选题策划立项、撰稿人周二至四成稿，审校为被动岗（受委托唤醒，演示成员间横向协作的唯一通道 delegation）。演示：周节奏流水线、被动岗受委托唤醒、文档工具面；编导段为 trigger(周一)→handoff 流水线（主编→策划→撰稿）+ 撰稿 dispatch 委托审校、审校 report 上行主编。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "topic-pitch-card", "description": "选题策划提交立项候选的标准格式：一句话、读者、角度、形态、交稿日。" }, { "name": "writing-standards", "description": "撰稿人的成稿纪律：钩子、段落结构、引用标注、完成度申报。" }, { "name": "review-checklist", "description": "审校岗的退稿/通过标准：事实、结构、用词三层核查，输出逐条理由。" } ], "mcpServers": [] } }, { "id": "EMGW14N8voa1b36LSuTxxw", "slug": "pack-emgw14n8voa1b36lsutxxw", "category": "pack", "marketCategory": "fleet", "tags": [], "canonical": { "brief": "官方团队模板：高频客服护航——客服组长每 4 小时带队巡检工单，一线客服处理、质检员抽检；组长对疑难工单用 escalate 合并上报（演示对抗每日防洪上限的上报纪律）与 CRM 客服工具面（凭据未配置时如实空转，不整体失败）。恰 3 岗满编无引用岗；编导段演示入站唤醒：event 边 bot:{{SERVICE_BOT}} 入站唤醒一线客服（客服 bot 为运营期配置的通道实例，缺位由运行期如实判定），一线 report 上行组长、组长可 dispatch 直接派单一线。铸造上架时 packId/agents/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "human": { "zh-CN": { "name": "客服班组", "summary": "官方团队模板：高频客服护航——客服组长每 4 小时带队巡检工单，一线客服处理、质检员抽检；组长对疑难工单用 escalate 合并上报（演示对抗每日防洪上限的上报纪律）与 CRM 客服工具面（凭据未配置时如实空转，不整体失败）。恰 3 岗满编无引用岗；编导段演示入站唤醒：event 边 bot:{{SERVICE_BOT}} 入站唤醒一线客服（客服 bot 为运营期配置的通道实例，缺位由运行期如实判定），一线 report 上行组长、组长可 dispatch 直接派单一线。铸造上架时 packId/agents/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "en": { "name": "客服班组", "summary": "官方团队模板：高频客服护航——客服组长每 4 小时带队巡检工单，一线客服处理、质检员抽检；组长对疑难工单用 escalate 合并上报（演示对抗每日防洪上限的上报纪律）与 CRM 客服工具面（凭据未配置时如实空转，不整体失败）。恰 3 岗满编无引用岗；编导段演示入站唤醒：event 边 bot:{{SERVICE_BOT}} 入站唤醒一线客服（客服 bot 为运营期配置的通道实例，缺位由运行期如实判定），一线 report 上行组长、组长可 dispatch 直接派单一线。铸造上架时 packId/agents/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "es": { "name": "客服班组", "summary": "官方团队模板：高频客服护航——客服组长每 4 小时带队巡检工单，一线客服处理、质检员抽检；组长对疑难工单用 escalate 合并上报（演示对抗每日防洪上限的上报纪律）与 CRM 客服工具面（凭据未配置时如实空转，不整体失败）。恰 3 岗满编无引用岗；编导段演示入站唤醒：event 边 bot:{{SERVICE_BOT}} 入站唤醒一线客服（客服 bot 为运营期配置的通道实例，缺位由运行期如实判定），一线 report 上行组长、组长可 dispatch 直接派单一线。铸造上架时 packId/agents/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "fr": { "name": "客服班组", "summary": "官方团队模板：高频客服护航——客服组长每 4 小时带队巡检工单，一线客服处理、质检员抽检；组长对疑难工单用 escalate 合并上报（演示对抗每日防洪上限的上报纪律）与 CRM 客服工具面（凭据未配置时如实空转，不整体失败）。恰 3 岗满编无引用岗；编导段演示入站唤醒：event 边 bot:{{SERVICE_BOT}} 入站唤醒一线客服（客服 bot 为运营期配置的通道实例，缺位由运行期如实判定），一线 report 上行组长、组长可 dispatch 直接派单一线。铸造上架时 packId/agents/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "ja": { "name": "客服班组", "summary": "官方团队模板：高频客服护航——客服组长每 4 小时带队巡检工单，一线客服处理、质检员抽检；组长对疑难工单用 escalate 合并上报（演示对抗每日防洪上限的上报纪律）与 CRM 客服工具面（凭据未配置时如实空转，不整体失败）。恰 3 岗满编无引用岗；编导段演示入站唤醒：event 边 bot:{{SERVICE_BOT}} 入站唤醒一线客服（客服 bot 为运营期配置的通道实例，缺位由运行期如实判定），一线 report 上行组长、组长可 dispatch 直接派单一线。铸造上架时 packId/agents/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "ticket-triage-five-steps", "description": "一线客服处理工单的标准分诊流程：读原文、分类、排序、答复、升级。" }, { "name": "cs-reply-etiquette", "description": "客服答复与升级件的标准措辞纪律：确认、方案、时限三段式与禁止项。" } ], "mcpServers": [] } }, { "id": "YMSt-hRNiNWXrlLYC8PSiA", "slug": "pack-ymst-hrninwxrllyc8psia", "category": "pack", "marketCategory": "fleet", "tags": [], "canonical": { "brief": "官方团队模板：三层级联的投研晨会——首席分析师汇总宏观/行业研究员晨报并拍板上行，风控官为被动岗（受委托或拍板批复唤醒）。演示：三层组织（{{fleetContext}} 只注入直接下级，中间层向上浓缩）、交易日节奏、escalate 驱动的被动岗；编导段为 trigger+handoff 级联（两研究员晨检完成接棒首席）与首席 dispatch 委托风控官。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "human": { "zh-CN": { "name": "投研晨会", "summary": "官方团队模板：三层级联的投研晨会——首席分析师汇总宏观/行业研究员晨报并拍板上行，风控官为被动岗（受委托或拍板批复唤醒）。演示：三层组织（{{fleetContext}} 只注入直接下级，中间层向上浓缩）、交易日节奏、escalate 驱动的被动岗；编导段为 trigger+handoff 级联（两研究员晨检完成接棒首席）与首席 dispatch 委托风控官。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "en": { "name": "Investment Research Morning Meeting", "summary": "官方团队模板：三层级联的投研晨会——首席分析师汇总宏观/行业研究员晨报并拍板上行，风控官为被动岗（受委托或拍板批复唤醒）。演示：三层组织（{{fleetContext}} 只注入直接下级，中间层向上浓缩）、交易日节奏、escalate 驱动的被动岗；编导段为 trigger+handoff 级联（两研究员晨检完成接棒首席）与首席 dispatch 委托风控官。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "es": { "name": "投研晨会", "summary": "官方团队模板：三层级联的投研晨会——首席分析师汇总宏观/行业研究员晨报并拍板上行，风控官为被动岗（受委托或拍板批复唤醒）。演示：三层组织（{{fleetContext}} 只注入直接下级，中间层向上浓缩）、交易日节奏、escalate 驱动的被动岗；编导段为 trigger+handoff 级联（两研究员晨检完成接棒首席）与首席 dispatch 委托风控官。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "fr": { "name": "投研晨会", "summary": "官方团队模板：三层级联的投研晨会——首席分析师汇总宏观/行业研究员晨报并拍板上行，风控官为被动岗（受委托或拍板批复唤醒）。演示：三层组织（{{fleetContext}} 只注入直接下级，中间层向上浓缩）、交易日节奏、escalate 驱动的被动岗；编导段为 trigger+handoff 级联（两研究员晨检完成接棒首席）与首席 dispatch 委托风控官。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "ja": { "name": "投研晨会", "summary": "官方团队模板：三层级联的投研晨会——首席分析师汇总宏观/行业研究员晨报并拍板上行，风控官为被动岗（受委托或拍板批复唤醒）。演示：三层组织（{{fleetContext}} 只注入直接下级，中间层向上浓缩）、交易日节奏、escalate 驱动的被动岗；编导段为 trigger+handoff 级联（两研究员晨检完成接棒首席）与首席 dispatch 委托风控官。铸造上架时 packId/agents 按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" } }, "whenToUse": "Browse under \"fleet\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "invest-view-card", "description": "研究员与首席的标准观点格式：观点、证据、反方、置信度四件套。" }, { "name": "morning-note-template", "description": "首席合成晨会产出的标准模板：共识、分歧、待拍板、观察清单。" } ], "mcpServers": [ { "registryName": "fd-daas-mcp" }, { "registryName": "fd-quant" } ] } }, { "id": "aofNf6o2LR61lPK8IVVkJw", "slug": "pack-aofnf6o2lr61lpk8ivvkjw", "category": "pack", "marketCategory": "cluster", "tags": [], "canonical": { "brief": "官方团队模板样板：CEO 巡视（{{fleetContext}} 注入）+ 三真数据岗（开源/投资）+ 助理引用岗 + 两连接器空态岗（电商/内容，演示待补岗降级）+ 萬星团队 a2a optional 成员（演示依赖缺位降级）。pack 自带 agents 上限 3（铸造实证），故助理为 existing-persona 引用岗；编导段为 supervisor 式 dispatch——CEO 可向各部门派活（显式 dispatch 边，非全舰 implicit）。铸造上架时 packId/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "human": { "zh-CN": { "name": "AI 公司", "summary": "官方团队模板样板：CEO 巡视（{{fleetContext}} 注入）+ 三真数据岗（开源/投资）+ 助理引用岗 + 两连接器空态岗（电商/内容，演示待补岗降级）+ 萬星团队 a2a optional 成员（演示依赖缺位降级）。pack 自带 agents 上限 3（铸造实证），故助理为 existing-persona 引用岗；编导段为 supervisor 式 dispatch——CEO 可向各部门派活（显式 dispatch 边，非全舰 implicit）。铸造上架时 packId/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "en": { "name": "AI 公司", "summary": "官方团队模板样板：CEO 巡视（{{fleetContext}} 注入）+ 三真数据岗（开源/投资）+ 助理引用岗 + 两连接器空态岗（电商/内容，演示待补岗降级）+ 萬星团队 a2a optional 成员（演示依赖缺位降级）。pack 自带 agents 上限 3（铸造实证），故助理为 existing-persona 引用岗；编导段为 supervisor 式 dispatch——CEO 可向各部门派活（显式 dispatch 边，非全舰 implicit）。铸造上架时 packId/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "es": { "name": "AI 公司", "summary": "官方团队模板样板：CEO 巡视（{{fleetContext}} 注入）+ 三真数据岗（开源/投资）+ 助理引用岗 + 两连接器空态岗（电商/内容，演示待补岗降级）+ 萬星团队 a2a optional 成员（演示依赖缺位降级）。pack 自带 agents 上限 3（铸造实证），故助理为 existing-persona 引用岗；编导段为 supervisor 式 dispatch——CEO 可向各部门派活（显式 dispatch 边，非全舰 implicit）。铸造上架时 packId/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "fr": { "name": "AI 公司", "summary": "官方团队模板样板：CEO 巡视（{{fleetContext}} 注入）+ 三真数据岗（开源/投资）+ 助理引用岗 + 两连接器空态岗（电商/内容，演示待补岗降级）+ 萬星团队 a2a optional 成员（演示依赖缺位降级）。pack 自带 agents 上限 3（铸造实证），故助理为 existing-persona 引用岗；编导段为 supervisor 式 dispatch——CEO 可向各部门派活（显式 dispatch 边，非全舰 implicit）。铸造上架时 packId/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" }, "ja": { "name": "AI 公司", "summary": "官方团队模板样板：CEO 巡视（{{fleetContext}} 注入）+ 三真数据岗（开源/投资）+ 助理引用岗 + 两连接器空态岗（电商/内容，演示待补岗降级）+ 萬星团队 a2a optional 成员（演示依赖缺位降级）。pack 自带 agents 上限 3（铸造实证），故助理为 existing-persona 引用岗；编导段为 supervisor 式 dispatch——CEO 可向各部门派活（显式 dispatch 边，非全舰 implicit）。铸造上架时 packId/占位符按铸造脚本补齐；本文件是清单真源，测试钉住其必须通过 fleet 段预校验。" } }, "whenToUse": "Browse under \"cluster\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "dept-patrol-report-framework", "description": "公司式舰队各部门向 CEO 汇报的标准框架：数字先行、异常归因、裁决事项。" }, { "name": "escalation-decision-card", "description": "CEO 与各部门提交升级事项的标准格式：背景、事实、选项与代价、建议。" } ], "mcpServers": [] } }, { "id": "50Tp176hW0HYTTWXZbarKA", "slug": "pack-50tp176hw0hyttwxzbarka", "category": "official-data", "marketCategory": "wire", "tags": [ "数据", "自助分析", "取数", "看板", "DAAS" ], "canonical": { "brief": "谦面自助分析消费八件套：取数（工作流清单/AkShare）、实体与集合管理、指标查询、看板检视、PDF 语义检索、头脑风暴与一键研究——业务人自助完成数据消费全流程" }, "human": { "zh-CN": { "name": "数据 · 自助分析", "summary": "谦面自助分析消费八件套：取数（工作流清单/AkShare）、实体与集合管理、指标查询、看板检视、PDF 语义检索、头脑风暴与一键研究——业务人自助完成数据消费全流程" }, "en": { "name": "数据 · 自助分析", "summary": "谦面自助分析消费八件套：取数（工作流清单/AkShare）、实体与集合管理、指标查询、看板检视、PDF 语义检索、头脑风暴与一键研究——业务人自助完成数据消费全流程" }, "es": { "name": "数据 · 自助分析", "summary": "谦面自助分析消费八件套：取数（工作流清单/AkShare）、实体与集合管理、指标查询、看板检视、PDF 语义检索、头脑风暴与一键研究——业务人自助完成数据消费全流程" }, "fr": { "name": "数据 · 自助分析", "summary": "谦面自助分析消费八件套：取数（工作流清单/AkShare）、实体与集合管理、指标查询、看板检视、PDF 语义检索、头脑风暴与一键研究——业务人自助完成数据消费全流程" }, "ja": { "name": "数据 · 自助分析", "summary": "谦面自助分析消费八件套：取数（工作流清单/AkShare）、实体与集合管理、指标查询、看板检视、PDF 语义检索、头脑风暴与一键研究——业务人自助完成数据消费全流程" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "fd-daas-based-data-fetch", "description": "按 data-fetch 工作流清单取数：经 fd-open-data-mcp 解析实体并按日期读概念值，返回数据行与来源标识，可选落库" }, { "name": "fd-daas-fetch-data", "description": "在 daas 注册表查实体、定位覆盖它的数据源，并在源数据上定义指标（daas_* 工具 + indicators 工作流清单）" }, { "name": "fd-datasource-akshare", "description": "用 scraw-akshare 抓 A 股行情与财务数据：全部 spider 命令、CLI 参数、库表结构与已知坑位速查，跑任何抓取前先读它" }, { "name": "fd-daas-research", "description": "跑完整研究管线（实体→指标→看板→报告）：经 research 工作流清单检索概念、落研究包并生成 markdown 报告" }, { "name": "fd-daas-brainstorm", "description": "研究目标头脑风暴：把模糊想法聊成具体研究计划（可锚定名家投资方法论），只产出计划再交棒研究管线" }, { "name": "fd-daas-dashboard", "description": "查找/打开/检视既有独立 HTML 看板：列出、按关键词搜索、看简介与数据血缘、浏览器打开并查询底层数据（只读）" }, { "name": "fd-daas-pdf", "description": "PDF/文本本地向量化与语义检索：抽取分块、sentence-transformers 嵌入存 daas.db，KNN 问答带文档页码引用，不出网" }, { "name": "fd-daas-entities-collection", "description": "实体与实体集日常管理：发现实体、集合增删改查、成员增删排序、审计历史与规则集合同步（不含新建规则）" } ], "mcpServers": [ { "registryName": "fd-daas-mcp" } ] } }, { "id": "oGURb41bdyAotVGOeBHZUA", "slug": "pack-ogurb41bdyaotvgoebhzua", "category": "official-data", "marketCategory": "wire", "tags": [ "数据", "指标", "创作", "规则", "DAAS" ], "canonical": { "brief": "谦面指标工坊创作五件套：规则驱动实体集、指标持久化、指标集策展、统一规则（json/script/position/llm）与单文件 HTML 看板——把业务口径沉淀为可复用的数据资产" }, "human": { "zh-CN": { "name": "数据 · 指标工坊", "summary": "谦面指标工坊创作五件套：规则驱动实体集、指标持久化、指标集策展、统一规则（json/script/position/llm）与单文件 HTML 看板——把业务口径沉淀为可复用的数据资产" }, "en": { "name": "数据 · 指标工坊", "summary": "谦面指标工坊创作五件套：规则驱动实体集、指标持久化、指标集策展、统一规则（json/script/position/llm）与单文件 HTML 看板——把业务口径沉淀为可复用的数据资产" }, "es": { "name": "数据 · 指标工坊", "summary": "谦面指标工坊创作五件套：规则驱动实体集、指标持久化、指标集策展、统一规则（json/script/position/llm）与单文件 HTML 看板——把业务口径沉淀为可复用的数据资产" }, "fr": { "name": "数据 · 指标工坊", "summary": "谦面指标工坊创作五件套：规则驱动实体集、指标持久化、指标集策展、统一规则（json/script/position/llm）与单文件 HTML 看板——把业务口径沉淀为可复用的数据资产" }, "ja": { "name": "数据 · 指标工坊", "summary": "谦面指标工坊创作五件套：规则驱动实体集、指标持久化、指标集策展、统一规则（json/script/position/llm）与单文件 HTML 看板——把业务口径沉淀为可复用的数据资产" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "fd-daas-entities-collection-creator", "description": "创建规则驱动的实体集（自选股/组合）：声明式 JSON 规则，跨表/动态逻辑落 Python 规则脚本可重跑" }, { "name": "fd-daas-indicators-creator", "description": "端到端建指标并持久化：扩展取数流程落 scraw_ 存储表并把序列数据存进表（刷新为手动重跑）" }, { "name": "fd-daas-indicators-collection-creator", "description": "策展指标集：把指标组成命名集合、按三级解析给出各成员 resolved score，导出机器可读 CSV + 人类可读简介" }, { "name": "fd-daas-rules-creator", "description": "统一规则创作（json/script/position/llm 四型）：daas_create_rule 落库、daas_test_rule 试跑，可挂实体/指标集合并同步" }, { "name": "fd-daas-dashboard-creator", "description": "把 daas 数据做成单文件独立 HTML 看板：ECharts + 实体/时间交互筛选，注册进 dashboards 表并迭代修改" } ], "mcpServers": [ { "registryName": "fd-daas-mcp" } ] } }, { "id": "INdsRWpgpp2Q03I0z3XoFA", "slug": "pack-indsrwpgpp2q03i0z3xofa", "category": "facade", "marketCategory": "operations", "tags": [ "assistant", "wanxing", "cluster-management" ], "canonical": { "brief": "在壹座中操作萬星机群的专用助手——浏览 agent 目录、读详情、查看部署状态。只读工具 + 提议通道，所有写操作需经动作闸确认。" }, "human": { "zh-CN": { "name": "萬星机群助手", "summary": "在壹座中操作萬星机群的专用助手——浏览 agent 目录、读详情、查看部署状态。只读工具 + 提议通道，所有写操作需经动作闸确认。" }, "en": { "name": "萬星机群助手", "summary": "在壹座中操作萬星机群的专用助手——浏览 agent 目录、读详情、查看部署状态。只读工具 + 提议通道，所有写操作需经动作闸确认。" }, "es": { "name": "萬星机群助手", "summary": "在壹座中操作萬星机群的专用助手——浏览 agent 目录、读详情、查看部署状态。只读工具 + 提议通道，所有写操作需经动作闸确认。" }, "fr": { "name": "萬星机群助手", "summary": "在壹座中操作萬星机群的专用助手——浏览 agent 目录、读详情、查看部署状态。只读工具 + 提议通道，所有写操作需经动作闸确认。" }, "ja": { "name": "萬星机群助手", "summary": "在壹座中操作萬星机群的专用助手——浏览 agent 目录、读详情、查看部署状态。只读工具 + 提议通道，所有写操作需经动作闸确认。" } }, "whenToUse": "Browse under \"operations\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "wanxing-browse-guidance", "description": "指导用户如何使用 wanxing-ops-mcp 的 browse/card 工具查看 agent" }, { "name": "deployment-precheck-guidance", "description": "指导用户在部署前检查凭据和机位状态" } ], "mcpServers": [ { "registryName": "wanxing-ops-mcp" } ] } }, { "id": "Ut_bJMuW5KxAD83KImPmlg", "slug": "pack-ut-bjmuw5kxad83kimpmlg", "category": "facade", "marketCategory": "facade", "tags": [ "分身", "企业接待", "寻数" ], "canonical": { "brief": "寻数（FindData）的企业 AI 分身：想了解我们的企业、在做的项目、产品具备的能力，直接和分身聊。回答出自随版本冻结的策展语料，每条带语料截至时间；答不了的如实说答不了，并建议找人对的人。" }, "human": { "zh-CN": { "name": "寻数分身", "summary": "寻数（FindData）的企业 AI 分身：想了解我们的企业、在做的项目、产品具备的能力，直接和分身聊。回答出自随版本冻结的策展语料，每条带语料截至时间；答不了的如实说答不了，并建议找人对的人。" }, "en": { "name": "寻数分身", "summary": "寻数（FindData）的企业 AI 分身：想了解我们的企业、在做的项目、产品具备的能力，直接和分身聊。回答出自随版本冻结的策展语料，每条带语料截至时间；答不了的如实说答不了，并建议找人对的人。" }, "es": { "name": "寻数分身", "summary": "寻数（FindData）的企业 AI 分身：想了解我们的企业、在做的项目、产品具备的能力，直接和分身聊。回答出自随版本冻结的策展语料，每条带语料截至时间；答不了的如实说答不了，并建议找人对的人。" }, "fr": { "name": "寻数分身", "summary": "寻数（FindData）的企业 AI 分身：想了解我们的企业、在做的项目、产品具备的能力，直接和分身聊。回答出自随版本冻结的策展语料，每条带语料截至时间；答不了的如实说答不了，并建议找人对的人。" }, "ja": { "name": "寻数分身", "summary": "寻数（FindData）的企业 AI 分身：想了解我们的企业、在做的项目、产品具备的能力，直接和分身聊。回答出自随版本冻结的策展语料，每条带语料截至时间；答不了的如实说答不了，并建议找人对的人。" } }, "whenToUse": "Browse under \"facade\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "corpus-company", "description": "企业语料：寻数是谁、双星一底座的产品定位与核心概念" }, { "name": "corpus-projects", "description": "项目语料：双星路线图里程碑状态、出口判据、在途变更" }, { "name": "corpus-capabilities", "description": "能力语料：谦面在架垂类、萬星服务形态、能拿来做什么" } ], "mcpServers": [] } }, { "id": "qCEIutJzQd49jnDtV4qeqw", "slug": "pack-qceiutjzqd49jndtv4qeqw", "category": "facade", "marketCategory": "facet", "tags": [ "assistant", "facet", "pack-authoring" ], "canonical": { "brief": "在壹座中操作谦面的专用助手——浏览包目录、读详情、发布前校验、建包向导。只读工具 + 提议通道，所有写操作需经动作闸确认。" }, "human": { "zh-CN": { "name": "谦面助手", "summary": "在壹座中操作谦面的专用助手——浏览包目录、读详情、发布前校验、建包向导。只读工具 + 提议通道，所有写操作需经动作闸确认。" }, "en": { "name": "谦面助手", "summary": "在壹座中操作谦面的专用助手——浏览包目录、读详情、发布前校验、建包向导。只读工具 + 提议通道，所有写操作需经动作闸确认。" }, "es": { "name": "谦面助手", "summary": "在壹座中操作谦面的专用助手——浏览包目录、读详情、发布前校验、建包向导。只读工具 + 提议通道，所有写操作需经动作闸确认。" }, "fr": { "name": "谦面助手", "summary": "在壹座中操作谦面的专用助手——浏览包目录、读详情、发布前校验、建包向导。只读工具 + 提议通道，所有写操作需经动作闸确认。" }, "ja": { "name": "谦面助手", "summary": "在壹座中操作谦面的专用助手——浏览包目录、读详情、发布前校验、建包向导。只读工具 + 提议通道，所有写操作需经动作闸确认。" } }, "whenToUse": "Browse under \"facet\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "facet-browse-guidance", "description": "指导用户如何使用 facet-market-mcp 的 browse/read 工具搜索包" }, { "name": "facet-validate-guidance", "description": "指导用户如何使用 validate 进行发布前预检" }, { "name": "facet-concepts", "description": "回答 facet 概念问题：pack 是什么、订阅与安装是两件事、版本快照语义" }, { "name": "facet-recommend", "description": "按用户需求推荐 pack：先问诊再检索，指向市场条目" }, { "name": "facet-permissions", "description": "解释 pack 的 MCP 引用与工具面（allowTools）披露口径" }, { "name": "facet-install-guidance", "description": "两条安装路径引导：壹座一键订阅安装 / CLI 安装到 Claude Code / Cursor" } ], "mcpServers": [ { "registryName": "facet-market-mcp" }, { "registryName": "facet-ops-mcp" } ] } }, { "id": "0aVynUNHmscTAxu7fGs-NQ", "slug": "pack-0avynunhmsctaxu7fgs-nq", "category": "pack", "marketCategory": "crm", "tags": [ "CRM", "客服", "Chatwoot" ], "canonical": { "brief": "商家自带 Chatwoot 的客服接待套件：连接你的 Chatwoot 账号后，读未处理会话、草拟回复、人工确认后发送，全程你自己的租户与凭据（BYOK）。含客服接待技能与使用指南。" }, "human": { "zh-CN": { "name": "CRM 客服套件", "summary": "商家自带 Chatwoot 的客服接待套件：连接你的 Chatwoot 账号后，读未处理会话、草拟回复、人工确认后发送，全程你自己的租户与凭据（BYOK）。含客服接待技能与使用指南。" }, "en": { "name": "CRM 客服套件", "summary": "商家自带 Chatwoot 的客服接待套件：连接你的 Chatwoot 账号后，读未处理会话、草拟回复、人工确认后发送，全程你自己的租户与凭据（BYOK）。含客服接待技能与使用指南。" }, "es": { "name": "CRM 客服套件", "summary": "商家自带 Chatwoot 的客服接待套件：连接你的 Chatwoot 账号后，读未处理会话、草拟回复、人工确认后发送，全程你自己的租户与凭据（BYOK）。含客服接待技能与使用指南。" }, "fr": { "name": "CRM 客服套件", "summary": "商家自带 Chatwoot 的客服接待套件：连接你的 Chatwoot 账号后，读未处理会话、草拟回复、人工确认后发送，全程你自己的租户与凭据（BYOK）。含客服接待技能与使用指南。" }, "ja": { "name": "CRM 客服套件", "summary": "商家自带 Chatwoot 的客服接待套件：连接你的 Chatwoot 账号后，读未处理会话、草拟回复、人工确认后发送，全程你自己的租户与凭据（BYOK）。含客服接待技能与使用指南。" } }, "whenToUse": "Browse under \"crm\" in the market.", "badges": { "official": true, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "crm-desk-reception-merchant", "description": "客服接待（商家自带 Chatwoot 租户）：发现账号、读未处理会话、草拟回复、确认后发送并复核；失败四分类如实上报。" } ], "mcpServers": [ { "registryName": "chatwoot" } ] } }, { "id": "Bzz9r5rmYPIAitS8Ca33Lw", "slug": "pack-bzz9r5rmypiaits8ca33lw", "category": "facade", "marketCategory": "wanxing", "tags": [ "萬星", "内容发布", "图文", "视频", "wanxing" ], "canonical": { "brief": "云端生成 + 本地发布的完整闭环：从 VAAS 托管实例拉取租户名下的生成产物（图/视频/图文），在本机完成平台适配、人工确认后经自己的浏览器登录态发布到 10 视频平台 + 11 图文平台。登录态只存本机、平台零保管（见包内 SECURITY-NOTES.md）。发布前必须人工确认。" }, "human": { "zh-CN": { "name": "萬星 · 本地内容发布套件", "summary": "云端生成 + 本地发布的完整闭环：从 VAAS 托管实例拉取租户名下的生成产物（图/视频/图文），在本机完成平台适配、人工确认后经自己的浏览器登录态发布到 10 视频平台 + 11 图文平台。登录态只存本机、平台零保管（见包内 SECURITY-NOTES.md）。发布前必须人工确认。" }, "en": { "name": "萬星 · 本地内容发布套件", "summary": "云端生成 + 本地发布的完整闭环：从 VAAS 托管实例拉取租户名下的生成产物（图/视频/图文），在本机完成平台适配、人工确认后经自己的浏览器登录态发布到 10 视频平台 + 11 图文平台。登录态只存本机、平台零保管（见包内 SECURITY-NOTES.md）。发布前必须人工确认。" }, "es": { "name": "萬星 · 本地内容发布套件", "summary": "云端生成 + 本地发布的完整闭环：从 VAAS 托管实例拉取租户名下的生成产物（图/视频/图文），在本机完成平台适配、人工确认后经自己的浏览器登录态发布到 10 视频平台 + 11 图文平台。登录态只存本机、平台零保管（见包内 SECURITY-NOTES.md）。发布前必须人工确认。" }, "fr": { "name": "萬星 · 本地内容发布套件", "summary": "云端生成 + 本地发布的完整闭环：从 VAAS 托管实例拉取租户名下的生成产物（图/视频/图文），在本机完成平台适配、人工确认后经自己的浏览器登录态发布到 10 视频平台 + 11 图文平台。登录态只存本机、平台零保管（见包内 SECURITY-NOTES.md）。发布前必须人工确认。" }, "ja": { "name": "萬星 · 本地内容发布套件", "summary": "云端生成 + 本地发布的完整闭环：从 VAAS 托管实例拉取租户名下的生成产物（图/视频/图文），在本机完成平台适配、人工确认后经自己的浏览器登录态发布到 10 视频平台 + 11 图文平台。登录态只存本机、平台零保管（见包内 SECURITY-NOTES.md）。发布前必须人工确认。" } }, "whenToUse": "Browse under \"wanxing\" in the market.", "badges": { "official": false, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "fd-vaas-pull-from-cloud", "description": "从 VAAS 托管生成实例拉取租户名下的产物到本机发布目录。三种输入任选其一：谦面调用键（连网关按 owner 过滤）、签名下载 URL（生成响应里带的）、或直接给 asset id + 实例地址。产物自动落 downloads/ 标准目录并生成发布技能可直接消费的 task 目录（article.md + meta.json 骨架）。无凭据时输出获取指引。**本技能只拉内容，不碰任何登录态**（发布登录走 fd-vaas-login，在本机完成）。" }, { "name": "fd-vaas-login", "description": "视频 + 图文平台登录管理页 — 查看各平台登录状态，触发扫码/窗口登录，轮询 QR 码。零依赖 Python stdlib HTTP Server。" }, { "name": "fd-vaas-publish-docs", "description": "把一篇文章/图文一键分发到主流图文平台(知乎、微信公众号、小红书、抖音图文、快手图文、雪球、东方财富号、同花顺财经号、今日头条、百家号、微博)。**只做编排 -- 平台差异化文案/字数/标签/封面 + 内容适配 + 发布记录**。**上游优先原则**:vendored social-auto-upload(在 fd-vaas-publish-videos skill 里)有图文(Note)实现的平台(小红书、抖音图文、快手图文)走 `scripts/note_adapter.py` 复用上游,cookie 与视频发布共享;上游没有的平台才用自己的逻辑:**默认 ego-browser**(r…" }, { "name": "fd-vaas-publish-videos", "description": ">" } ], "mcpServers": [ { "registryName": "vaas-wanxing" } ] } }, { "id": "0RdQE184anPUaMhIvFwndA", "slug": "pack-0rdqe184anpuamhivfwnda", "category": "facade", "marketCategory": "facet", "tags": [ "谦面", "市场", "发布", "MCP", "运维" ], "canonical": { "brief": "用 AI 直接操作谦面市场与网关：浏览/检索功集与网关条目、发布新功集与新版本、上下架、切可见性、策展分类、发布前查重。装包即得 facet_market_* 与 facet_agent_* 十二个工具（凭据走壹座既有连接：registry + 谦面市场配对），配两份手册技能讲清『怎么写 manifest、怎么过发布门、怎么安全写操作』。" }, "human": { "zh-CN": { "name": "谦面市场助手", "summary": "用 AI 直接操作谦面市场与网关：浏览/检索功集与网关条目、发布新功集与新版本、上下架、切可见性、策展分类、发布前查重。装包即得 facet_market_* 与 facet_agent_* 十二个工具（凭据走壹座既有连接：registry + 谦面市场配对），配两份手册技能讲清『怎么写 manifest、怎么过发布门、怎么安全写操作』。" }, "en": { "name": "谦面市场助手", "summary": "用 AI 直接操作谦面市场与网关：浏览/检索功集与网关条目、发布新功集与新版本、上下架、切可见性、策展分类、发布前查重。装包即得 facet_market_* 与 facet_agent_* 十二个工具（凭据走壹座既有连接：registry + 谦面市场配对），配两份手册技能讲清『怎么写 manifest、怎么过发布门、怎么安全写操作』。" }, "es": { "name": "谦面市场助手", "summary": "用 AI 直接操作谦面市场与网关：浏览/检索功集与网关条目、发布新功集与新版本、上下架、切可见性、策展分类、发布前查重。装包即得 facet_market_* 与 facet_agent_* 十二个工具（凭据走壹座既有连接：registry + 谦面市场配对），配两份手册技能讲清『怎么写 manifest、怎么过发布门、怎么安全写操作』。" }, "fr": { "name": "谦面市场助手", "summary": "用 AI 直接操作谦面市场与网关：浏览/检索功集与网关条目、发布新功集与新版本、上下架、切可见性、策展分类、发布前查重。装包即得 facet_market_* 与 facet_agent_* 十二个工具（凭据走壹座既有连接：registry + 谦面市场配对），配两份手册技能讲清『怎么写 manifest、怎么过发布门、怎么安全写操作』。" }, "ja": { "name": "谦面市场助手", "summary": "用 AI 直接操作谦面市场与网关：浏览/检索功集与网关条目、发布新功集与新版本、上下架、切可见性、策展分类、发布前查重。装包即得 facet_market_* 与 facet_agent_* 十二个工具（凭据走壹座既有连接：registry + 谦面市场配对），配两份手册技能讲清『怎么写 manifest、怎么过发布门、怎么安全写操作』。" } }, "whenToUse": "Browse under \"facet\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "facet-market-publishing", "description": "谦面功集发布实操手册：manifest 写法、校验门、工具声明门、配额与版本纪律，发布失败逐条排障。" }, { "name": "facet-gateway-ops", "description": "谦面网关（registry）操作手册：AgentCard 目录浏览/登记/更新/启停/查重的安全用法与错误语义。" } ], "mcpServers": [ { "registryName": "facet-market-mcp" }, { "registryName": "facet-ops-mcp" } ] } }, { "id": "ev5E51eth7e5McvyDrl9Rg", "slug": "pack-ev5e51eth7e5mcvydrl9rg", "category": "facade", "marketCategory": "wire", "tags": [ "market", "briefing", "速览" ], "canonical": { "brief": "把一段财经新闻、公告或行情评论收敛成一页速览：三行结论、五个要点、一处存疑。为需要在盘前/会前快速消化的读者而生。" }, "human": { "zh-CN": { "name": "市场快讯 · 一页速览", "summary": "把一段财经新闻、公告或行情评论收敛成一页速览：三行结论、五个要点、一处存疑。为需要在盘前/会前快速消化的读者而生。" }, "en": { "name": "市场快讯 · 一页速览", "summary": "把一段财经新闻、公告或行情评论收敛成一页速览：三行结论、五个要点、一处存疑。为需要在盘前/会前快速消化的读者而生。" }, "es": { "name": "市场快讯 · 一页速览", "summary": "把一段财经新闻、公告或行情评论收敛成一页速览：三行结论、五个要点、一处存疑。为需要在盘前/会前快速消化的读者而生。" }, "fr": { "name": "市场快讯 · 一页速览", "summary": "把一段财经新闻、公告或行情评论收敛成一页速览：三行结论、五个要点、一处存疑。为需要在盘前/会前快速消化的读者而生。" }, "ja": { "name": "市场快讯 · 一页速览", "summary": "把一段财经新闻、公告或行情评论收敛成一页速览：三行结论、五个要点、一处存疑。为需要在盘前/会前快速消化的读者而生。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "market-flash-briefing", "description": "把一段财经新闻/公告/评论收敛成一页速览：三行结论、五个要点、一处存疑，输出固定版式。" } ], "mcpServers": [] } }, { "id": "tHd65-7SfBT_z0KfTvHGIw", "slug": "pack-thd65-7sfbt-z0kftvhgiw", "category": "facade", "marketCategory": "facet", "tags": [ "文档" ], "canonical": { "brief": "M4③ 套装验收——第二件外作者绿地带货" }, "human": { "zh-CN": { "name": "外部好货 · 简报快写", "summary": "M4③ 套装验收——第二件外作者绿地带货" }, "en": { "name": "外部好货 · 简报快写", "summary": "M4③ 套装验收——第二件外作者绿地带货" }, "es": { "name": "外部好货 · 简报快写", "summary": "M4③ 套装验收——第二件外作者绿地带货" }, "fr": { "name": "外部好货 · 简报快写", "summary": "M4③ 套装验收——第二件外作者绿地带货" }, "ja": { "name": "外部好货 · 简报快写", "summary": "M4③ 套装验收——第二件外作者绿地带货" } }, "whenToUse": "Browse under \"facet\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "brief-draft", "description": "简报占位" } ], "mcpServers": [] } }, { "id": "6kDsKgmb-3_86A81VZC6UA", "slug": "pack-6kdskgmb-3-86a81vzc6ua", "category": "facade", "marketCategory": "facet", "tags": [ "数据" ], "canonical": { "brief": "M4② 绿地从目录部署验收——外作者的可部署 pack" }, "human": { "zh-CN": { "name": "外部好货 · 数据速览", "summary": "M4② 绿地从目录部署验收——外作者的可部署 pack" }, "en": { "name": "外部好货 · 数据速览", "summary": "M4② 绿地从目录部署验收——外作者的可部署 pack" }, "es": { "name": "外部好货 · 数据速览", "summary": "M4② 绿地从目录部署验收——外作者的可部署 pack" }, "fr": { "name": "外部好货 · 数据速览", "summary": "M4② 绿地从目录部署验收——外作者的可部署 pack" }, "ja": { "name": "外部好货 · 数据速览", "summary": "M4② 绿地从目录部署验收——外作者的可部署 pack" } }, "whenToUse": "Browse under \"facet\" in the market.", "badges": { "official": true, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "quick-look", "description": "速览占位" } ], "mcpServers": [] } }, { "id": "Aqag8Mky4ksoPmwRDMCVYw", "slug": "pack-aqag8mky4ksopmwrdmcvyw", "category": "facade", "marketCategory": "wire", "tags": [ "官方功能集", "生态", "数据" ], "canonical": { "brief": "官方功能集（生态首发）：FindData 数据底座（fd-open-data-mcp 宏观/金融真数据）+ 官方 time/fetch 生态 server + 数据研究 persona。装上即可做可溯源的数据问答与网页取证。" }, "human": { "zh-CN": { "name": "数据研究 · 生态版", "summary": "官方功能集（生态首发）：FindData 数据底座（fd-open-data-mcp 宏观/金融真数据）+ 官方 time/fetch 生态 server + 数据研究 persona。装上即可做可溯源的数据问答与网页取证。" }, "en": { "name": "数据研究 · 生态版", "summary": "官方功能集（生态首发）：FindData 数据底座（fd-open-data-mcp 宏观/金融真数据）+ 官方 time/fetch 生态 server + 数据研究 persona。装上即可做可溯源的数据问答与网页取证。" }, "es": { "name": "数据研究 · 生态版", "summary": "官方功能集（生态首发）：FindData 数据底座（fd-open-data-mcp 宏观/金融真数据）+ 官方 time/fetch 生态 server + 数据研究 persona。装上即可做可溯源的数据问答与网页取证。" }, "fr": { "name": "数据研究 · 生态版", "summary": "官方功能集（生态首发）：FindData 数据底座（fd-open-data-mcp 宏观/金融真数据）+ 官方 time/fetch 生态 server + 数据研究 persona。装上即可做可溯源的数据问答与网页取证。" }, "ja": { "name": "数据研究 · 生态版", "summary": "官方功能集（生态首发）：FindData 数据底座（fd-open-data-mcp 宏观/金融真数据）+ 官方 time/fetch 生态 server + 数据研究 persona。装上即可做可溯源的数据问答与网页取证。" } }, "whenToUse": "Browse under \"wire\" in the market.", "badges": { "official": false, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "eco-skill-creator", "description": "[生态] 结构化创建新技能包（anthropics/skills，Apache-2.0）" } ], "mcpServers": [ { "registryName": "fd-open-data-mcp" }, { "registryName": "eco-mcp-time" }, { "registryName": "eco-mcp-fetch" } ] } }, { "id": "qnhd7B3J1-HKDDKzZQ_fUQ", "slug": "pack-qnhd7b3j1-hkddkzzq-fuq", "category": "facade", "marketCategory": "doc-studio", "tags": [ "官方功能集", "文档", "Office", "报告" ], "canonical": { "brief": "Office 文档生成与编辑：商务报告（docx）、数据工作簿（xlsx）、汇报演示（pptx）三场景成稿，配验证过的封面配方、中式设计系统与生成后质检 postcheck；支持编辑用户上传的原件（fetch_document_file 取原件）。回合内免费。" }, "human": { "zh-CN": { "name": "文档工坊", "summary": "Office 文档生成与编辑：商务报告（docx）、数据工作簿（xlsx）、汇报演示（pptx）三场景成稿，配验证过的封面配方、中式设计系统与生成后质检 postcheck；支持编辑用户上传的原件（fetch_document_file 取原件）。回合内免费。" }, "en": { "name": "文档工坊", "summary": "Office 文档生成与编辑：商务报告（docx）、数据工作簿（xlsx）、汇报演示（pptx）三场景成稿，配验证过的封面配方、中式设计系统与生成后质检 postcheck；支持编辑用户上传的原件（fetch_document_file 取原件）。回合内免费。" }, "es": { "name": "文档工坊", "summary": "Office 文档生成与编辑：商务报告（docx）、数据工作簿（xlsx）、汇报演示（pptx）三场景成稿，配验证过的封面配方、中式设计系统与生成后质检 postcheck；支持编辑用户上传的原件（fetch_document_file 取原件）。回合内免费。" }, "fr": { "name": "文档工坊", "summary": "Office 文档生成与编辑：商务报告（docx）、数据工作簿（xlsx）、汇报演示（pptx）三场景成稿，配验证过的封面配方、中式设计系统与生成后质检 postcheck；支持编辑用户上传的原件（fetch_document_file 取原件）。回合内免费。" }, "ja": { "name": "文档工坊", "summary": "Office 文档生成与编辑：商务报告（docx）、数据工作簿（xlsx）、汇报演示（pptx）三场景成稿，配验证过的封面配方、中式设计系统与生成后质检 postcheck；支持编辑用户上传的原件（fetch_document_file 取原件）。回合内免费。" } }, "whenToUse": "Browse under \"doc-studio\" in the market.", "badges": { "official": false, "facade": true }, "provenance": { "source": "https://facet.finddatatech.cloud", "updatedAt": "2026-11-02" }, "contents": { "skills": [ { "name": "doc-studio", "description": "文档工坊——Word/Excel/PPT 商务文档生成与编辑。当用户要求生成报告、表格、演示文稿、文档交付物，或要求编辑/转换上传的 docx/xlsx/pptx 文件时使用。产出符合中式商务版式的成稿，生成后自动质检。" } ], "mcpServers": [] } }, { "id": "skill:contract-review", "slug": "skill-contract-review", "category": "skill", "marketCategory": null, "tags": [ "法律", "合同审查" ], "canonical": { "brief": "Chinese contract review: systematically reviews commercial contract clauses across risk dimensions and outputs a structured report with ready-to-use revised clauses." }, "human": { "zh-CN": { "name": "contract-review", "summary": "中文合同审查:风险维度系统审查商事合同条款,输出结构化报告与可直接使用的修改条款" }, "en": { "name": "contract-review", "summary": "Chinese contract review: systematically reviews commercial contract clauses across risk dimensions and outputs a structured report with ready-to-use revised clauses." }, "es": { "name": "contract-review", "summary": "Revisión de contratos en chino: revisa sistemáticamente las cláusulas de contratos comerciales por dimensiones de riesgo y genera un informe estructurado con cláusulas modificadas listas para usar." }, "fr": { "name": "contract-review", "summary": "Révision de contrats en chinois : examine systématiquement les clauses des contrats commerciaux selon des dimensions de risque et produit un rapport structuré avec des clauses modifiées prêtes à l'emploi." }, "ja": { "name": "contract-review", "summary": "中国語契約審査：リスクの観点から商事契約の条項を体系的に審査し、構造化されたレポートとそのまま使える修正条項を出力します。" } }, "whenToUse": "Relevant when working on: 法律, 合同审查.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:financial-statement-analysis", "slug": "skill-financial-statement-analysis", "category": "skill", "marketCategory": null, "tags": [ "金融", "财报分析" ], "canonical": { "brief": "Listed-company financial statement analysis: links the three statements, performs ratio analysis and identifies red flags of financial window dressing, producing a buy-side research report." }, "human": { "zh-CN": { "name": "financial-statement-analysis", "summary": "上市公司财报分析:三表联动+比率分析+财务粉饰红旗识别,输出买方视角研究报告" }, "en": { "name": "financial-statement-analysis", "summary": "Listed-company financial statement analysis: links the three statements, performs ratio analysis and identifies red flags of financial window dressing, producing a buy-side research report." }, "es": { "name": "financial-statement-analysis", "summary": "Análisis de estados financieros de sociedades cotizadas: vincula los tres estados, realiza análisis de ratios e identifica señales de alerta de maquillaje financiero, generando un informe de investigación con perspectiva compradora." }, "fr": { "name": "financial-statement-analysis", "summary": "Analyse des états financiers de sociétés cotées : relie les trois états, effectue une analyse par ratios et identifie les signaux d'alerte de window dressing financier, pour produire un rapport de recherche du point de vue acheteur." }, "ja": { "name": "financial-statement-analysis", "summary": "上場企業の財務諸表分析：3表の連動、比率分析、財務粉飾のレッドフラッグ識別を行い、バイサイド視点のリサーチレポートを出力します。" } }, "whenToUse": "Relevant when working on: 金融, 财报分析.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:legal-research-cn", "slug": "skill-legal-research-cn", "category": "skill", "marketCategory": null, "tags": [ "法律", "法律检索" ], "canonical": { "brief": "Chinese legal research and argumentation: statute retrieval with validity verification, matching of constitutive elements, and adversarial argumentation, covering civil, commercial, labor, and corporate fields." }, "human": { "zh-CN": { "name": "legal-research-cn", "summary": "中国法律检索与论证:法条检索时效校验+构成要件匹配+对抗性论证,民商事劳动公司领域" }, "en": { "name": "legal-research-cn", "summary": "Chinese legal research and argumentation: statute retrieval with validity verification, matching of constitutive elements, and adversarial argumentation, covering civil, commercial, labor, and corporate fields." }, "es": { "name": "legal-research-cn", "summary": "Investigación y argumentación jurídica en China: búsqueda de normas con verificación de vigencia, coincidencia de elementos constitutivos y argumentación contradictoria, en los ámbitos civil, mercantil, laboral y societario." }, "fr": { "name": "legal-research-cn", "summary": "Recherche juridique et argumentation en droit chinois : recherche de textes avec vérification de validité, appariement des éléments constitutifs et argumentation contradictoire, dans les domaines civil, commercial, du travail et des sociétés." }, "ja": { "name": "legal-research-cn", "summary": "中国法のリサーチと論証：法令検索と有効性の検証、構成要件のマッチング、対抗的論証を提供し、民事・商事・労働・会社法分野を対象とします。" } }, "whenToUse": "Relevant when working on: 法律, 法律检索.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:earnings-analysis", "slug": "skill-earnings-analysis", "category": "skill", "marketCategory": null, "tags": [ "finance", "research-methods", "finddata", "anthropics" ], "canonical": { "brief": "Create professional equity research earnings update reports (8-12 pages, 3,000-5,000 words) analyzing quarterly results for companies already under coverage. Fast-turnaround format focusing on beat/miss analysis, key metrics, updated estimates, and revised thesis. Includes 1-3 summary tables and 8-1" }, "human": { "zh-CN": { "name": "earnings-analysis", "summary": "Create professional equity research earnings update reports (8-12 pages, 3,000-5,000 words) analyzing quarterly results for companies already under coverage. Fast-turnaround format focusing on beat/miss analysis, key metrics, updated estimates, and revised thesis. Includes 1-3 summary tables and 8-12 charts. Use when user requests \"earnings update\", \"quarterly update\", \"earnings analysis\", \"Q1/Q2/Q3/Q4 results\", or post-earnings report." }, "en": { "name": "earnings-analysis", "summary": "Create professional equity research earnings update reports (8-12 pages, 3,000-5,000 words) analyzing quarterly results for companies already under coverage. Fast-turnaround format focusing on beat/miss analysis, key metrics, updated estimates, and revised thesis. Includes 1-3 summary tables and 8-12 charts. Use when user requests \"earnings update\", \"quarterly update\", \"earnings analysis\", \"Q1/Q2/Q3/Q4 results\", or post-earnings report." }, "es": { "name": "earnings-analysis", "summary": "Cree informes profesionales de actualización de resultados de research de renta variable (8-12 páginas, 3.000-5.000 palabras) que analizan los resultados trimestrales de empresas ya bajo cobertura. Formato de rápida respuesta centrado en el análisis beat/miss, las métricas clave, las estimaciones actualizadas y la tesis revisada. Incluye de 1 a 3 tablas resumen y de 8 a 12 gráficos. Úselo cuando el usuario solicite una «actualización de resultados», una «actualización trimestral», un «análisis de resultados», los «resultados del Q1/Q2/Q3/Q4» o un informe posterior a resultados." }, "fr": { "name": "earnings-analysis", "summary": "Créez des rapports professionnels d'actualisation de résultats d'equity research (8-12 pages, 3 000-5 000 mots) analysant les résultats trimestriels de sociétés déjà sous couverture. Format à rotation rapide centré sur l'analyse beat/miss, les indicateurs clés, les estimations actualisées et la thèse révisée. Comprend 1 à 3 tableaux récapitulatifs et 8 à 12 graphiques. À utiliser lorsque l'utilisateur demande une « mise à jour de résultats », une « mise à jour trimestrielle », une « analyse de résultats », les « résultats du T1/T2/T3/T4 » ou un rapport post-résultats." }, "ja": { "name": "earnings-analysis", "summary": "すでにカバー対象となっている企業の四半期業績を分析する、プロフェッショナルな決算アップデートレポート（8〜12ページ、3,000〜5,000語）を作成します。beat/miss分析、主要指標、改訂後の推計、修正された投資テーゼに焦点を当てた速報形式です。サマリーテーブル1〜3個とチャート8〜12点を含みます。ユーザーが「決算アップデート」「四半期アップデート」「決算分析」「Q1/Q2/Q3/Q4の結果」を求めた場合や、決算後レポートを依頼した場合に使用します。" } }, "whenToUse": "Relevant when working on: finance, research-methods, finddata, anthropics.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/research-methods/earnings-analysis" }, { "id": "skill:sector-overview", "slug": "skill-sector-overview", "category": "skill", "marketCategory": null, "tags": [ "finance", "research-methods", "finddata", "anthropics" ], "canonical": { "brief": "Create comprehensive industry and sector landscape reports covering market dynamics, competitive positioning, key players, and thematic trends. Use for client requests, sector initiations, thematic research pieces, or internal knowledge building. Triggers on \"sector overview\", \"industry report\", \"ma" }, "human": { "zh-CN": { "name": "sector-overview", "summary": "Create comprehensive industry and sector landscape reports covering market dynamics, competitive positioning, key players, and thematic trends. Use for client requests, sector initiations, thematic research pieces, or internal knowledge building. Triggers on \"sector overview\", \"industry report\", \"market landscape\", \"sector analysis\", \"industry deep dive\", or \"thematic research\"." }, "en": { "name": "sector-overview", "summary": "Create comprehensive industry and sector landscape reports covering market dynamics, competitive positioning, key players, and thematic trends. Use for client requests, sector initiations, thematic research pieces, or internal knowledge building. Triggers on \"sector overview\", \"industry report\", \"market landscape\", \"sector analysis\", \"industry deep dive\", or \"thematic research\"." }, "es": { "name": "sector-overview", "summary": "Cree informes completos del panorama industrial y sectorial que cubran la dinámica del mercado, el posicionamiento competitivo, los actores clave y las tendencias temáticas. Úselo para solicitudes de clientes, inicios de cobertura de un sector, piezas de investigación temática o construcción de conocimiento interno. Se activa con «resumen sectorial», «informe industrial», «panorama de mercado», «análisis sectorial», «análisis en profundidad de la industria» o «investigación temática»." }, "fr": { "name": "sector-overview", "summary": "Créez des rapports complets sur le paysage industriel et sectoriel couvrant la dynamique du marché, le positionnement concurrentiel, les acteurs clés et les tendances thématiques. À utiliser pour les demandes clients, les lancements de couverture sectorielle, les études thématiques ou la constitution de connaissances internes. Se déclenche sur « aperçu du secteur », « rapport sectoriel », « paysage du marché », « analyse sectorielle », « analyse approfondie d'une industrie » ou « recherche thématique »." }, "ja": { "name": "sector-overview", "summary": "市場ダイナミクス、競争ポジション、主要プレイヤー、テーマ別トレンドを網羅した産業・セクター全体像レポートを作成します。クライアントの依頼、セクターの新規カバー、テーマ別リサーチ、社内ナレッジの構築に使用します。「セクター概観」「業界レポート」「市場ランドスケープ」「セクター分析」「業界深掘り」「テーマ別リサーチ」で起動します。" } }, "whenToUse": "Relevant when working on: finance, research-methods, finddata, anthropics.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/research-methods/sector-overview" }, { "id": "skill:comps-analysis", "slug": "skill-comps-analysis", "category": "skill", "marketCategory": null, "tags": [ "finance", "research-methods", "finddata", "anthropics" ], "canonical": { "brief": "Build institutional-grade comparable company analyses with operating metrics, valuation multiples, and statistical benchmarking in Excel/spreadsheet format." }, "human": { "zh-CN": { "name": "comps-analysis", "summary": "Build institutional-grade comparable company analyses with operating metrics, valuation multiples, and statistical benchmarking in Excel/spreadsheet format." }, "en": { "name": "comps-analysis", "summary": "Build institutional-grade comparable company analyses with operating metrics, valuation multiples, and statistical benchmarking in Excel/spreadsheet format." }, "es": { "name": "comps-analysis", "summary": "Cree análisis de empresas comparables de nivel institucional con métricas operativas, múltiplos de valoración y benchmarking estadístico en formato Excel/hoja de cálculo." }, "fr": { "name": "comps-analysis", "summary": "Réalisez des analyses de sociétés comparables de qualité institutionnelle avec des métriques opérationnelles, des multiples de valorisation et un benchmarking statistique, au format Excel/feuille de calcul." }, "ja": { "name": "comps-analysis", "summary": "Excel/スプレッドシート形式で、事業指標、バリュエーション倍率、統計ベンチマーキングを備えた機関投資家水準のコンプス分析を作成します。" } }, "whenToUse": "Relevant when working on: finance, research-methods, finddata, anthropics.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/research-methods/comps-analysis" }, { "id": "skill:dcf-model", "slug": "skill-dcf-model", "category": "skill", "marketCategory": null, "tags": [ "finance", "research-methods", "finddata", "anthropics" ], "canonical": { "brief": "Real DCF (Discounted Cash Flow) model creation for equity valuation. Retrieves financial data from SEC filings and analyst reports, builds comprehensive cash flow projections with proper WACC calculations, performs sensitivity analysis, and outputs professional Excel models with executive summaries." }, "human": { "zh-CN": { "name": "dcf-model", "summary": "Real DCF (Discounted Cash Flow) model creation for equity valuation. Retrieves financial data from SEC filings and analyst reports, builds comprehensive cash flow projections with proper WACC calculations, performs sensitivity analysis, and outputs professional Excel models with executive summaries. Use when users need to value a company using DCF methodology, request intrinsic value analysis, or ask for detailed financial modeling with growth projections and terminal value calculations." }, "en": { "name": "dcf-model", "summary": "Real DCF (Discounted Cash Flow) model creation for equity valuation. Retrieves financial data from SEC filings and analyst reports, builds comprehensive cash flow projections with proper WACC calculations, performs sensitivity analysis, and outputs professional Excel models with executive summaries. Use when users need to value a company using DCF methodology, request intrinsic value analysis, or ask for detailed financial modeling with growth projections and terminal value calculations." }, "es": { "name": "dcf-model", "summary": "Creación real de modelos DCF (flujo de caja descontado) para valoración de renta variable. Obtiene datos financieros de filings de la SEC y de reportes de analistas, construye proyecciones integrales de flujo de caja con cálculos adecuados del WACC, realiza análisis de sensibilidad y genera modelos profesionales de Excel con resúmenes ejecutivos. Úselo cuando el usuario necesite valorar una empresa con la metodología DCF, solicite un análisis de valor intrínseco o pida un modelado financiero detallado con proyecciones de crecimiento y cálculos de valor terminal." }, "fr": { "name": "dcf-model", "summary": "Création de véritables modèles DCF (flux de trésorerie actualisés) pour la valorisation d'actions. Récupère les données financières des filings SEC et des rapports d'analystes, construit des projections complètes de flux de trésorerie avec des calculs appropriés du WACC, effectue une analyse de sensibilité et produit des modèles Excel professionnels avec résumés exécutifs. À utiliser lorsque l'utilisateur doit valoriser une société selon la méthodologie DCF, demande une analyse de valeur intrinsèque ou requiert une modélisation financière détaillée avec projections de croissance et calculs de valeur terminale." }, "ja": { "name": "dcf-model", "summary": "株式バリュエーションのための本格的なDCF（割引キャッシュフロー）モデルを作成します。SEC提出書類やアナリストレポートから財務データを取得し、適切なWACC計算を用いた包括的なキャッシュフロー予測を構築し、感応度分析を実行し、エグゼクティブサマリー付きのプロフェッショナルなExcelモデルを出力します。DCF手法で企業を評価したい場合、内在価値分析を依頼された場合、成長予測とターミナルバリューの計算を含む詳細な財務モデリングが必要な場合に使用します。" } }, "whenToUse": "Relevant when working on: finance, research-methods, finddata, anthropics.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/research-methods/dcf-model" }, { "id": "skill:3-statement-model", "slug": "skill-3-statement-model", "category": "skill", "marketCategory": null, "tags": [ "finance", "research-methods", "finddata", "anthropics" ], "canonical": { "brief": "Complete, populate and fill out 3-statement financial model templates (Income Statement, Balance Sheet, Cash Flow Statement) . Use when asked to fill out model templates, complete existing model frameworks, populate financial models with data, complete a partially filled IS/BS/CF framework, or link " }, "human": { "zh-CN": { "name": "3-statement-model", "summary": "Complete, populate and fill out 3-statement financial model templates (Income Statement, Balance Sheet, Cash Flow Statement) . Use when asked to fill out model templates, complete existing model frameworks, populate financial models with data, complete a partially filled IS/BS/CF framework, or link integrated financial statements within an existing template structure. Triggers include requests to fill in, complete, or populate a 3-statement model template" }, "en": { "name": "3-statement-model", "summary": "Complete, populate and fill out 3-statement financial model templates (Income Statement, Balance Sheet, Cash Flow Statement) . Use when asked to fill out model templates, complete existing model frameworks, populate financial models with data, complete a partially filled IS/BS/CF framework, or link integrated financial statements within an existing template structure. Triggers include requests to fill in, complete, or populate a 3-statement model template" }, "es": { "name": "3-statement-model", "summary": "Complete y rellene plantillas de modelos financieros de 3 estados (estado de resultados, balance de situación, estado de flujos de caja). Úselo cuando se le pida rellenar plantillas de modelos, completar marcos de modelos existentes, poblar modelos financieros con datos, completar un marco IS/BS/CF parcialmente rellenado o vincular estados financieros integrados dentro de una estructura de plantilla existente. Los disparadores incluyen solicitudes de rellenar, completar o poblar una plantilla de modelo de 3 estados" }, "fr": { "name": "3-statement-model", "summary": "Complétez et remplissez des modèles de modèles financiers à 3 états (compte de résultat, bilan, tableau de flux de trésorerie). À utiliser lorsqu'il est demandé de remplir des modèles de modèles, de compléter des structures de modèles existantes, d'alimenter des modèles financiers avec des données, de compléter un cadre IS/BS/CF partiellement rempli ou de lier des états financiers intégrés dans une structure de modèle existante. Les déclencheurs incluent les demandes de remplir, compléter ou alimenter un modèle à 3 états" }, "ja": { "name": "3-statement-model", "summary": "3表財務モデルのテンプレート（損益計算書、貸借対照表、キャッシュフロー計算書）の完成・入力・記入を行います。モデルテンプレートの記入、既存モデルフレームの完成、財務モデルへのデータ入力、一部未記入のIS/BS/CFフレームの完成、既存テンプレート構造内での財務諸表のリンクを依頼された場合に使用します。トリガーには、3表モデルテンプレートの記入・完成・入力を求めるリクエストが含まれます" } }, "whenToUse": "Relevant when working on: finance, research-methods, finddata, anthropics.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/research-methods/3-statement-model" }, { "id": "skill:returns-analysis", "slug": "skill-returns-analysis", "category": "skill", "marketCategory": null, "tags": [ "finance", "research-methods", "finddata", "anthropics" ], "canonical": { "brief": "Build quick IRR/MOIC sensitivity tables for PE deal evaluation. Models returns across entry multiple, leverage, exit multiple, growth, and hold period scenarios. Use when sizing up a deal, stress-testing assumptions, or preparing IC returns exhibits. Triggers on \"returns analysis\", \"IRR sensitivity\"" }, "human": { "zh-CN": { "name": "returns-analysis", "summary": "Build quick IRR/MOIC sensitivity tables for PE deal evaluation. Models returns across entry multiple, leverage, exit multiple, growth, and hold period scenarios. Use when sizing up a deal, stress-testing assumptions, or preparing IC returns exhibits. Triggers on \"returns analysis\", \"IRR sensitivity\", \"MOIC table\", \"what's the return at\", \"model the returns\", or \"back of the envelope\"." }, "en": { "name": "returns-analysis", "summary": "Build quick IRR/MOIC sensitivity tables for PE deal evaluation. Models returns across entry multiple, leverage, exit multiple, growth, and hold period scenarios. Use when sizing up a deal, stress-testing assumptions, or preparing IC returns exhibits. Triggers on \"returns analysis\", \"IRR sensitivity\", \"MOIC table\", \"what's the return at\", \"model the returns\", or \"back of the envelope\"." }, "es": { "name": "returns-analysis", "summary": "Cree tablas rápidas de sensibilidad de TIR/MOIC para evaluar operaciones de PE. Modela los rendimientos según escenarios de múltiplo de entrada, apalancamiento, múltiplo de salida, crecimiento y periodo de tenencia. Úselo al dimensionar una operación, someter supuestos a pruebas de estrés o preparar anexos de rentabilidad para el IC. Se activa con «análisis de rentabilidad», «sensibilidad de TIR», «tabla de MOIC», «¿cuál es el retorno en…?», «modela los rendimientos» o «a bote pronto»." }, "fr": { "name": "returns-analysis", "summary": "Créez des tableaux de sensibilité TRI/MOIC rapides pour l'évaluation de deals de PE. Modélise les rendements selon des scénarios de multiple d'entrée, de levier, de multiple de sortie, de croissance et de durée de détention. À utiliser pour évaluer la taille d'un deal, tester la robustesse des hypothèses ou préparer des annexes de rendements pour l'IC. Se déclenche sur « analyse des rendements », « sensibilité du TRI », « tableau MOIC », « quel est le rendement à », « modélise les rendements » ou « sur un coin de table »." }, "ja": { "name": "returns-analysis", "summary": "PEディール評価のためのIRR/MOIC感応度テーブルを素早く作成します。エントリーマルチプル、レバレッジ、エグジットマルチプル、成長率、保有期間の各シナリオにわたってリターンをモデル化します。ディールの規模感の把握、前提条件のストレステスト、IC向けリターン資料の準備に使用します。「リターン分析」「IRR感応度」「MOICテーブル」「この条件でのリターンは」「リターンをモデル化して」「概算で」で起動します。" } }, "whenToUse": "Relevant when working on: finance, research-methods, finddata, anthropics.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/research-methods/returns-analysis" }, { "id": "skill:ic-memo", "slug": "skill-ic-memo", "category": "skill", "marketCategory": null, "tags": [ "finance", "research-methods", "finddata", "anthropics" ], "canonical": { "brief": "Draft a structured investment committee memo for PE deal approval. Synthesizes due diligence findings, financial analysis, and deal terms into a professional IC-ready document. Use when preparing for investment committee, writing up a deal, or creating a formal recommendation. Triggers on \"write IC " }, "human": { "zh-CN": { "name": "ic-memo", "summary": "Draft a structured investment committee memo for PE deal approval. Synthesizes due diligence findings, financial analysis, and deal terms into a professional IC-ready document. Use when preparing for investment committee, writing up a deal, or creating a formal recommendation. Triggers on \"write IC memo\", \"investment committee memo\", \"deal write-up\", \"prepare IC materials\", or \"recommendation memo\"." }, "en": { "name": "ic-memo", "summary": "Draft a structured investment committee memo for PE deal approval. Synthesizes due diligence findings, financial analysis, and deal terms into a professional IC-ready document. Use when preparing for investment committee, writing up a deal, or creating a formal recommendation. Triggers on \"write IC memo\", \"investment committee memo\", \"deal write-up\", \"prepare IC materials\", or \"recommendation memo\"." }, "es": { "name": "ic-memo", "summary": "Redacte un memorando estructurado para el comité de inversiones con vistas a la aprobación de una operación de PE. Sintetiza los hallazgos de la due diligence, el análisis financiero y los términos de la operación en un documento profesional listo para el IC. Úselo al preparar el comité de inversiones, redactar una operación o crear una recomendación formal. Se activa con «escribir memorando IC», «memorando del comité de inversiones», «informe de la operación», «preparar materiales para el IC» o «memorando de recomendación»." }, "fr": { "name": "ic-memo", "summary": "Rédigez une note structurée pour le comité d'investissement en vue de l'approbation d'un deal de PE. Synthétise les conclusions de la due diligence, l'analyse financière et les termes du deal dans un document professionnel prêt pour l'IC. À utiliser pour préparer un comité d'investissement, rédiger un deal ou établir une recommandation formelle. Se déclenche sur « rédiger une note IC », « note du comité d'investissement », « write-up du deal », « préparer les supports IC » ou « note de recommandation »." }, "ja": { "name": "ic-memo", "summary": "PEディールの承認に向けて、投資委員会向けの構造化されたメモを作成します。デューデリジェンスの調査結果、財務分析、ディール条件を、ICにそのまま提出できるプロフェッショナルなドキュメントにまとめます。投資委員会の準備、ディールの取りまとめ、正式な推奨案の作成に使用します。「ICメモを作成」「投資委員会メモ」「ディールの取りまとめ」「IC資料を準備」「推奨メモ」で起動します。" } }, "whenToUse": "Relevant when working on: finance, research-methods, finddata, anthropics.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/research-methods/ic-memo" }, { "id": "skill:macro-rates-monitor", "slug": "skill-macro-rates-monitor", "category": "skill", "marketCategory": null, "tags": [ "finance", "research-methods", "finddata", "anthropics", "lseg" ], "canonical": { "brief": "Build macroeconomic and rates dashboards combining macro indicators, yield curves, inflation breakevens, and swap rates. Use when monitoring macro conditions, analyzing yield curve shape, decomposing real vs nominal rates, assessing policy rate expectations, or evaluating financial conditions." }, "human": { "zh-CN": { "name": "macro-rates-monitor", "summary": "Build macroeconomic and rates dashboards combining macro indicators, yield curves, inflation breakevens, and swap rates. Use when monitoring macro conditions, analyzing yield curve shape, decomposing real vs nominal rates, assessing policy rate expectations, or evaluating financial conditions." }, "en": { "name": "macro-rates-monitor", "summary": "Build macroeconomic and rates dashboards combining macro indicators, yield curves, inflation breakevens, and swap rates. Use when monitoring macro conditions, analyzing yield curve shape, decomposing real vs nominal rates, assessing policy rate expectations, or evaluating financial conditions." }, "es": { "name": "macro-rates-monitor", "summary": "Cree paneles macroeconómicos y de tipos que combinen indicadores macro, curvas de tipos, breakevens de inflación y tipos de swaps. Úselo para monitorear condiciones macroeconómicas, analizar la forma de la curva de tipos, descomponer tipos reales frente a nominales, evaluar expectativas de tipos de política o valorar condiciones financieras." }, "fr": { "name": "macro-rates-monitor", "summary": "Créez des tableaux de bord macroéconomiques et de taux combinant indicateurs macro, courbes de taux, breakevens d'inflation et taux de swaps. À utiliser pour surveiller les conditions macroéconomiques, analyser la forme de la courbe des taux, décomposer taux réels et nominaux, évaluer les attentes de taux directeur ou apprécier les conditions financières." }, "ja": { "name": "macro-rates-monitor", "summary": "マクロ指標、イールドカーブ、インフレ・ブレークイーブン、スワップ金利を組み合わせた、マクロ・金利ダッシュボードを作成します。マクロ環境のモニタリング、イールドカーブの形状分析、実質金利と名目金利の分解、政策金利の期待の評価、金融環境の評価に使用します。" } }, "whenToUse": "Relevant when working on: finance, research-methods, finddata, anthropics, lseg.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/research-methods/macro-rates-monitor" }, { "id": "skill:company-valuation", "slug": "skill-company-valuation", "category": "skill", "marketCategory": null, "tags": [ "finance", "market-analysis", "finddata", "himself65" ], "canonical": { "brief": "Estimate the intrinsic value of a public company using DCF, relative (peer multiple) and sum-of-parts (SOTP) methods, then triangulate to an implied share price with upside/downside versus the current market price. Use this skill whenever the user asks: \"what is AAPL worth\", \"valuation of NVDA\", \"fa" }, "human": { "zh-CN": { "name": "company-valuation", "summary": "Estimate the intrinsic value of a public company using DCF, relative (peer multiple) and sum-of-parts (SOTP) methods, then triangulate to an implied share price with upside/downside versus the current market price. Use this skill whenever the user asks: \"what is AAPL worth\", \"valuation of NVDA\", \"fair value of TSLA\", \"intrinsic value\", \"DCF for MSFT\", \"build a DCF\", \"discounted cash flow\", \"WACC\", \"terminal value\", \"implied share price\", \"upside to fair value\", \"is X overvalued/undervalued\", \"relative valuation\", \"peer comparison valuation\", \"EV/EBITDA target\", \"SOTP\", \"sum of the parts\", \"how much is [company] worth\", \"price target from fundamentals\", \"value this company\", or any ticker in the context of computing intrinsic or relative valuation. Default to running ALL three methods (DCF + relative + SOTP-if-applicable) and presenting a blended implied price with a sensitivity table. Do not answer valuation questions from memory — always run the workflow." }, "en": { "name": "company-valuation", "summary": "Estimate the intrinsic value of a public company using DCF, relative (peer multiple) and sum-of-parts (SOTP) methods, then triangulate to an implied share price with upside/downside versus the current market price. Use this skill whenever the user asks: \"what is AAPL worth\", \"valuation of NVDA\", \"fair value of TSLA\", \"intrinsic value\", \"DCF for MSFT\", \"build a DCF\", \"discounted cash flow\", \"WACC\", \"terminal value\", \"implied share price\", \"upside to fair value\", \"is X overvalued/undervalued\", \"relative valuation\", \"peer comparison valuation\", \"EV/EBITDA target\", \"SOTP\", \"sum of the parts\", \"how much is [company] worth\", \"price target from fundamentals\", \"value this company\", or any ticker in the context of computing intrinsic or relative valuation. Default to running ALL three methods (DCF + relative + SOTP-if-applicable) and presenting a blended implied price with a sensitivity table. Do not answer valuation questions from memory — always run the workflow." }, "es": { "name": "company-valuation", "summary": "Estime el valor intrínseco de una sociedad cotizada mediante métodos DCF, relativo (múltiplos de comparables) y de suma de partes (SOTP), y luego triangule hacia un precio implícito por acción con upside/downside frente al precio actual de mercado. Use esta habilidad siempre que el usuario pregunte: «cuánto vale AAPL», «valoración de NVDA», «valor justo de TSLA», «valor intrínseco», «DCF para MSFT», «construir un DCF», «flujo de caja descontado», «WACC», «valor terminal», «precio implícito por acción», «upside hasta el valor justo», «¿está X sobrevalorada/infravalorada?», «valoración relativa», «valoración por comparables», «objetivo de EV/EBITDA», «SOTP», «suma de las partes», «cuánto vale [empresa]», «precio objetivo a partir de los fundamentales», «valora esta empresa», o cualquier ticker en el contexto de calcular una valoración intrínseca o relativa. Por defecto, ejecute LOS TRES métodos (DCF + relativo + SOTP si aplica) y presente un precio implícito combinado con una tabla de sensibilidad. No responda preguntas de valoración de memoria: ejecute siempre el flujo de trabajo." }, "fr": { "name": "company-valuation", "summary": "Estimez la valeur intrinsèque d'une société cotée à l'aide des méthodes DCF, relative (multiples de pairs) et de somme des parties (SOTP), puis triangulez vers un prix implicite par action avec l'upside/downside par rapport au prix de marché actuel. Utilisez cette compétence chaque fois que l'utilisateur demande : « combien vaut AAPL », « valorisation de NVDA », « juste valeur de TSLA », « valeur intrinsèque », « DCF pour MSFT », « construire un DCF », « flux de trésorerie actualisés », « WACC », « valeur terminale », « prix implicite par action », « upside vers la juste valeur », « X est-elle surévaluée/sous-évaluée », « valorisation relative », « valorisation par comparables », « cible de EV/EBITDA », « SOTP », « somme des parties », « combien vaut [société] », « prix cible fondé sur les fondamentaux », « valorise cette société », ou tout ticker dans le contexte d'un calcul de valorisation intrinsèque ou relative. Par défaut, exécutez LES TROIS méthodes (DCF + relative + SOTP si applicable) et présentez un prix implicite mixte avec un tableau de sensibilité. Ne répondez pas aux questions de valorisation de mémoire : exécutez toujours le workflow." }, "ja": { "name": "company-valuation", "summary": "DCF、相対バリエーション（類似企業倍率）、SOTP（部分合計法）の各手法により上場企業の内在価値を推定し、現在の市場価格に対するアップサイド/ダウンサイド付きのインプライド株価へと三角測量します。ユーザーが「AAPLはいくらの価値があるか」「NVDAのバリュエーション」「TSLAのフェアバリュー」「内在価値」「MSFTのDCF」「DCFを構築」「割引キャッシュフロー」「WACC」「ターミナルバリュー」「インプライド株価」「フェアバリューまでのアップサイド」「Xは過大評価/過小評価か」「相対バリエーション」「類似企業比較バリュエーション」「EV/EBITDAターゲット」「SOTP」「部分の合計」「[企業]の価値はいくら」「ファンダメンタルズからの目標株価」「この会社を評価して」など、内在価値や相対バリエーションの計算の文脈で任意のティッカーについて尋ねた場合に、このスキルを使用します。デフォルトでは3つの手法すべて（DCF＋相対バリエーション＋該当すればSOTP）を実行し、感応度テーブル付きのブレンドされたインプライド株価を提示します。バリュエーションの質問には記憶で答えず、必ずワークフローを実行してください。" } }, "whenToUse": "Relevant when working on: finance, market-analysis, finddata, himself65.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/market-analysis/company-valuation" }, { "id": "skill:stock-correlation", "slug": "skill-stock-correlation", "category": "skill", "marketCategory": null, "tags": [ "finance", "market-analysis", "finddata", "himself65" ], "canonical": { "brief": "Analyze stock correlations to find related companies and trading pairs. Use when the user asks about correlated stocks, related companies, sector peers, trading pairs, or how two or more stocks move together. Triggers: \"what correlates with NVDA\", \"find stocks related to AMD\", \"correlation between A" }, "human": { "zh-CN": { "name": "stock-correlation", "summary": "Analyze stock correlations to find related companies and trading pairs. Use when the user asks about correlated stocks, related companies, sector peers, trading pairs, or how two or more stocks move together. Triggers: \"what correlates with NVDA\", \"find stocks related to AMD\", \"correlation between AAPL and MSFT\", \"what moves with\", \"sector peers\", \"pair trading\", \"correlated stocks\", \"when NVDA drops what else drops\", \"stocks that move together\", \"beta to\", \"relative performance\", \"supply chain partners\", \"correlation matrix\", \"co-movement\", \"related tickers\", \"sympathy plays\", \"semiconductor peers\", \"hedging pair\", \"realized correlation\", \"rolling correlation\", or any request about stocks that move in tandem or inversely. Also triggers for well-known pairs like AMD/NVDA, GOOGL/AVGO, LITE/COHR. If only one ticker is provided, infer the user wants correlated peers." }, "en": { "name": "stock-correlation", "summary": "Analyze stock correlations to find related companies and trading pairs. Use when the user asks about correlated stocks, related companies, sector peers, trading pairs, or how two or more stocks move together. Triggers: \"what correlates with NVDA\", \"find stocks related to AMD\", \"correlation between AAPL and MSFT\", \"what moves with\", \"sector peers\", \"pair trading\", \"correlated stocks\", \"when NVDA drops what else drops\", \"stocks that move together\", \"beta to\", \"relative performance\", \"supply chain partners\", \"correlation matrix\", \"co-movement\", \"related tickers\", \"sympathy plays\", \"semiconductor peers\", \"hedging pair\", \"realized correlation\", \"rolling correlation\", or any request about stocks that move in tandem or inversely. Also triggers for well-known pairs like AMD/NVDA, GOOGL/AVGO, LITE/COHR. If only one ticker is provided, infer the user wants correlated peers." }, "es": { "name": "stock-correlation", "summary": "Analice correlaciones entre acciones para encontrar empresas relacionadas y pares de trading. Úselo cuando el usuario pregunte por acciones correlacionadas, empresas relacionadas, comparables del sector, pares de trading o cómo se mueven juntas dos o más acciones. Disparadores: «qué correlaciona con NVDA», «encuentra acciones relacionadas con AMD», «correlación entre AAPL y MSFT», «qué se mueve con», «comparables del sector», «pair trading», «acciones correlacionadas», «cuando NVDA cae qué más cae», «acciones que se mueven juntas», «beta respecto a», «rendimiento relativo», «socios de la cadena de suministro», «matriz de correlación», «co-movimiento», «tickers relacionados», «jugadas de simpatía», «comparables de semiconductores», «par de cobertura», «correlación realizada», «correlación móvil», o cualquier solicitud sobre acciones que se muevan en tándem o a la inversa. También se activa para pares conocidos como AMD/NVDA, GOOGL/AVGO, LITE/COHR. Si solo se facilita un ticker, infiera que el usuario quiere comparables correlacionados." }, "fr": { "name": "stock-correlation", "summary": "Analysez les corrélations entre actions pour trouver des sociétés liées et des paires de trading. À utiliser lorsque l'utilisateur s'interroge sur des actions corrélées, des sociétés liées, des pairs sectoriels, des paires de trading ou la façon dont deux actions ou plus évoluent ensemble. Déclencheurs : « qu'est-ce qui est corrélé à NVDA », « trouver des actions liées à AMD », « corrélation entre AAPL et MSFT », « qu'est-ce qui bouge avec », « pairs sectoriels », « pair trading », « actions corrélées », « quand NVDA baisse quoi d'autre baisse », « actions qui évoluent ensemble », « bêta par rapport à », « performance relative », « partenaires de la chaîne d'approvisionnement », « matrice de corrélation », « co-mouvement », « tickers liés », « trades de sympathie », « pairs des semi-conducteurs », « paire de couverture », « corrélation réalisée », « corrélation glissante », ou toute demande concernant des actions évoluant de concert ou à l'inverse. Se déclenche aussi pour des paires connues comme AMD/NVDA, GOOGL/AVGO, LITE/COHR. Si un seul ticker est fourni, déduisez que l'utilisateur veut des pairs corrélés." }, "ja": { "name": "stock-correlation", "summary": "株価相関を分析して、関連企業やペアトレードの候補を見つけます。ユーザーが相関のある銘柄、関連企業、セクター内の類似銘柄、ペアトレード、2銘柄以上の連動について尋ねた場合に使用します。トリガー：「NVDAと相関があるのは何か」「AMDに関連する銘柄を探す」「AAPLとMSFTの相関」「何と連動するか」「セクター内の類似銘柄」「ペアトレード」「相関のある銘柄」「NVDAが下がるとき他に何が下がるか」「一緒に動く銘柄」「〜に対するベータ」「相対パフォーマンス」「サプライチェーンパートナー」「相関マトリクス」「連動」「関連ティッカー」「シンパシープレイ」「半導体の類似銘柄」「ヘッジペア」「実現相関」「移動相関」、または銘柄の正逆の連動に関するあらゆるリクエスト。AMD/NVDA、GOOGL/AVGO、LITE/COHRなどの有名ペアでも起動します。ティッカーが1つだけ指定された場合は、相関のある類似銘柄を求めていると判断します。" } }, "whenToUse": "Relevant when working on: finance, market-analysis, finddata, himself65.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/market-analysis/stock-correlation" }, { "id": "skill:earnings-preview", "slug": "skill-earnings-preview", "category": "skill", "marketCategory": null, "tags": [ "finance", "market-analysis", "finddata", "himself65" ], "canonical": { "brief": "Generate a pre-earnings briefing for any stock using Yahoo Finance data. Use this skill whenever the user wants to prepare for an upcoming earnings report, understand what analysts expect, review a company's beat/miss track record, or get a quick overview before an earnings call. Triggers include: \"" }, "human": { "zh-CN": { "name": "earnings-preview", "summary": "Generate a pre-earnings briefing for any stock using Yahoo Finance data. Use this skill whenever the user wants to prepare for an upcoming earnings report, understand what analysts expect, review a company's beat/miss track record, or get a quick overview before an earnings call. Triggers include: \"earnings preview for AAPL\", \"what to expect from TSLA earnings\", \"MSFT reports next week\", \"earnings preview\", \"pre-earnings analysis\", \"what are analysts expecting for NVDA\", \"earnings estimates for\", \"will GOOGL beat earnings\", \"earnings beat/miss history\", \"upcoming earnings\", \"before earnings\", \"earnings setup\", \"consensus estimates\", \"earnings whisper\", \"EPS expectations\", \"what's the street expecting\", \"earnings season preview\", any mention of preparing for or previewing an earnings report, or any request to understand expectations ahead of a company's earnings date. Always use this skill when the user mentions a ticker in context of upcoming earnings, even if they don't say \"preview\" " }, "en": { "name": "earnings-preview", "summary": "Generate a pre-earnings briefing for any stock using Yahoo Finance data. Use this skill whenever the user wants to prepare for an upcoming earnings report, understand what analysts expect, review a company's beat/miss track record, or get a quick overview before an earnings call. Triggers include: \"earnings preview for AAPL\", \"what to expect from TSLA earnings\", \"MSFT reports next week\", \"earnings preview\", \"pre-earnings analysis\", \"what are analysts expecting for NVDA\", \"earnings estimates for\", \"will GOOGL beat earnings\", \"earnings beat/miss history\", \"upcoming earnings\", \"before earnings\", \"earnings setup\", \"consensus estimates\", \"earnings whisper\", \"EPS expectations\", \"what's the street expecting\", \"earnings season preview\", any mention of preparing for or previewing an earnings report, or any request to understand expectations ahead of a company's earnings date. Always use this skill when the user mentions a ticker in context of upcoming earnings, even if they don't say \"preview\" " }, "es": { "name": "earnings-preview", "summary": "Genere un informe previo a resultados para cualquier acción usando datos de Yahoo Finance. Use esta habilidad cuando el usuario quiera prepararse para un próximo informe de resultados, entender qué esperan los analistas, revisar el historial de beat/miss de una empresa u obtener una visión rápida antes de la conferencia de resultados. Los disparadores incluyen: «previa de resultados de AAPL», «qué esperar de los resultados de TSLA», «MSFT publica la próxima semana», «previa de resultados», «análisis previo a resultados», «qué esperan los analistas para NVDA», «estimaciones de resultados de», «¿superará GOOGL los resultados?», «historial de beat/miss de resultados», «próximos resultados», «antes de resultados», «setup de resultados», «estimaciones de consenso», «earnings whisper», «expectativas de EPS», «qué espera el mercado», «avance de la temporada de resultados», cualquier mención a la preparación o anticipación de un informe de resultados, o cualquier solicitud de entender las expectativas antes de la fecha de resultados de una empresa. Use siempre esta habilidad cuando el usuario mencione un ticker en el contexto de próximos resultados, aunque no diga «previa»." }, "fr": { "name": "earnings-preview", "summary": "Générez un briefing pré-résultats pour toute action à partir des données Yahoo Finance. Utilisez cette compétence lorsque l'utilisateur veut se préparer à un prochain rapport de résultats, comprendre les attentes des analystes, passer en revue l'historique beat/miss d'une société ou obtenir un aperçu rapide avant une conférence de résultats. Déclencheurs : « aperçu des résultats d'AAPL », « à quoi s'attendre des résultats de TSLA », « MSFT publie la semaine prochaine », « aperçu des résultats », « analyse pré-résultats », « qu'attendent les analystes pour NVDA », « estimations de résultats pour », « GOOGL battra-t-il les résultats », « historique beat/miss des résultats », « résultats à venir », « avant les résultats », « setup de résultats », « estimations du consensus », « earnings whisper », « attentes d'EPS », « qu'attend la place », « aperçu de la saison des résultats », toute mention de préparation ou d'anticipation d'un rapport de résultats, ou toute demande visant à comprendre les attentes avant la date de résultats d'une société. Utilisez toujours cette compétence lorsque l'utilisateur mentionne un ticker dans le contexte de résultats à venir, même s'il ne dit pas « aperçu »." }, "ja": { "name": "earnings-preview", "summary": "Yahoo Financeのデータを使って、あらゆる銘柄の決算前ブリーフィングを作成します。ユーザーが今後の決算発表に備えたい、アナリストの予想を知りたい、企業のbeat/miss実績を確認したい、決算説明会の前に概要を把握したいといった場合に使用します。トリガー：「AAPLの決算プレビュー」「TSLAの決算で何を期待すべきか」「MSFTが来週発表」「決算プレビュー」「決算前分析」「NVDAでアナリストは何を予想しているか」「〜の決算推計」「GOOGLは決算で上振れるか」「決算beat/miss履歴」「今後の決算」「決算前に」「決算セットアップ」「コンセンサス推計」「earnings whisper」「EPS予想」「市場の予想は」「決算シーズン予習」、決算発表の準備や予習に関するあらゆる言及、企業の決算発表日前に期待を知りたいというあらゆるリクエスト。ユーザーが今後の決算の文脈でティッカーに言及した場合は、「プレビュー」と言わなくても必ずこのスキルを使用します。" } }, "whenToUse": "Relevant when working on: finance, market-analysis, finddata, himself65.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/market-analysis/earnings-preview" }, { "id": "skill:earnings-recap", "slug": "skill-earnings-recap", "category": "skill", "marketCategory": null, "tags": [ "finance", "market-analysis", "finddata", "himself65" ], "canonical": { "brief": "Generate a post-earnings analysis for any stock using Yahoo Finance data. Use when the user wants to review what happened after earnings, understand beat/miss results, see stock reaction, or get an earnings recap. Triggers: \"AAPL earnings recap\", \"how did TSLA earnings go\", \"MSFT earnings results\", " }, "human": { "zh-CN": { "name": "earnings-recap", "summary": "Generate a post-earnings analysis for any stock using Yahoo Finance data. Use when the user wants to review what happened after earnings, understand beat/miss results, see stock reaction, or get an earnings recap. Triggers: \"AAPL earnings recap\", \"how did TSLA earnings go\", \"MSFT earnings results\", \"did NVDA beat earnings\", \"post-earnings analysis\", \"earnings surprise\", \"what happened with GOOGL earnings\", \"earnings reaction\", \"stock moved after earnings\", \"EPS beat or miss\", \"revenue beat or miss\", \"quarterly results for\", \"how were earnings\", \"AMZN reported last night\", \"earnings call recap\", or any request about a company's recent earnings outcome. Use this skill when the user references a past earnings event, even if they just say \"AAPL reported\" or \"how did they do\"." }, "en": { "name": "earnings-recap", "summary": "Generate a post-earnings analysis for any stock using Yahoo Finance data. Use when the user wants to review what happened after earnings, understand beat/miss results, see stock reaction, or get an earnings recap. Triggers: \"AAPL earnings recap\", \"how did TSLA earnings go\", \"MSFT earnings results\", \"did NVDA beat earnings\", \"post-earnings analysis\", \"earnings surprise\", \"what happened with GOOGL earnings\", \"earnings reaction\", \"stock moved after earnings\", \"EPS beat or miss\", \"revenue beat or miss\", \"quarterly results for\", \"how were earnings\", \"AMZN reported last night\", \"earnings call recap\", or any request about a company's recent earnings outcome. Use this skill when the user references a past earnings event, even if they just say \"AAPL reported\" or \"how did they do\"." }, "es": { "name": "earnings-recap", "summary": "Genere un análisis posterior a resultados para cualquier acción usando datos de Yahoo Finance. Úselo cuando el usuario quiera repasar lo ocurrido tras los resultados, entender los beat/miss, ver la reacción de la acción u obtener un resumen de resultados. Disparadores: «resumen de resultados de AAPL», «cómo fueron los resultados de TSLA», «resultados de MSFT», «¿superó NVDA los resultados?», «análisis post-resultados», «sorpresa de resultados», «qué pasó con los resultados de GOOGL», «reacción a los resultados», «la acción se movió tras los resultados», «beat o miss de EPS», «beat o miss de ingresos», «resultados trimestrales de», «cómo fueron los resultados», «AMZN publicó anoche», «resumen de la conferencia de resultados», o cualquier solicitud sobre el resultado reciente de los resultados de una empresa. Use esta habilidad cuando el usuario se refiera a un evento de resultados pasado, aunque solo diga «AAPL publicó» o «cómo les fue»." }, "fr": { "name": "earnings-recap", "summary": "Générez une analyse post-résultats pour toute action à partir des données Yahoo Finance. À utiliser lorsque l'utilisateur veut passer en revue ce qui s'est passé après les résultats, comprendre les beat/miss, voir la réaction du titre ou obtenir un récapitulatif des résultats. Déclencheurs : « récapitulatif des résultats d'AAPL », « comment se sont passés les résultats de TSLA », « résultats de MSFT », « NVDA a-t-il battu les résultats », « analyse post-résultats », « surprise de résultats », « que s'est-il passé avec les résultats de GOOGL », « réaction aux résultats », « le titre a bougé après les résultats », « EPS beat ou miss », « chiffre d'affaires beat ou miss », « résultats trimestriels de », « comment ont été les résultats », « AMZN a publié hier soir », « récapitulatif de la conférence de résultats », ou toute demande concernant les résultats récents d'une société. Utilisez cette compétence lorsque l'utilisateur évoque un événement de résultats passé, même s'il dit simplement « AAPL a publié » ou « comment ont-ils fait »." }, "ja": { "name": "earnings-recap", "summary": "Yahoo Financeのデータを使って、あらゆる銘柄の決算後分析を作成します。決算後の状況を振り返りたい、beat/missの結果を知りたい、株価反応を確認したい、決算まとめが欲しいという場合に使用します。トリガー：「AAPLの決算まとめ」「TSLAの決算はどうだった」「MSFTの決算結果」「NVDAは決算で上振れたか」「決算後分析」「決算サプライズ」「GOOGLの決算で何が起きたか」「決算への反応」「決算後に株価が動いた」「EPSのbeat/miss」「売上のbeat/miss」「〜の四半期業績」「決算はどうだった」「AMZNが昨晩発表」「決算説明会のまとめ」など、企業の直近の決算結果に関するあらゆるリクエスト。過去の決算イベントに言及した場合には、「AAPLが発表した」「業績はどうだった」という言い方でもこのスキルを使用します。" } }, "whenToUse": "Relevant when working on: finance, market-analysis, finddata, himself65.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/market-analysis/earnings-recap" }, { "id": "skill:akshare-stock", "slug": "skill-akshare-stock", "category": "skill", "marketCategory": null, "tags": [ "finance", "cn-market", "a-share", "finddata", "molezzz" ], "canonical": { "brief": "All-in-one A-share analysis skill (real-time quotes, technicals, fundamentals, sectors, derivatives, and cross-market), built on akshare with natural-language routing." }, "human": { "zh-CN": { "name": "akshare-stock", "summary": "A股分析全能 Skill（实时行情、技术面、基本面、板块、衍生品与跨市场），基于 akshare + 自然语言路由" }, "en": { "name": "akshare-stock", "summary": "All-in-one A-share analysis skill (real-time quotes, technicals, fundamentals, sectors, derivatives, and cross-market), built on akshare with natural-language routing." }, "es": { "name": "akshare-stock", "summary": "Habilidad todo en uno para el análisis de acciones A (cotizaciones en tiempo real, análisis técnico, fundamentales, sectores, derivados y mercados cruzados), basada en akshare con enrutamiento por lenguaje natural." }, "fr": { "name": "akshare-stock", "summary": "Compétence tout-en-un d'analyse des actions A (cotations en temps réel, analyse technique, fondamentaux, secteurs, produits dérivés et marchés croisés), basée sur akshare avec routage en langage naturel." }, "ja": { "name": "akshare-stock", "summary": "A株分析のオールインワンスキル（リアルタイム相場、テクニカル、ファンダメンタルズ、セクター、デリバティブ、クロスマーケット）。akshareベースで、自然言語ルーティングに対応しています。" } }, "whenToUse": "Relevant when working on: finance, cn-market, a-share, finddata, molezzz.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/cn-market/akshare-stock" }, { "id": "skill:stock-analysis", "slug": "skill-stock-analysis", "category": "skill", "marketCategory": null, "tags": [ "finance", "us-market", "finddata", "claude-office-skills" ], "canonical": { "brief": "Analyze stocks with fundamental and technical analysis. Supports US, China A-shares, and Hong Kong markets. Generate investment reports with key metrics." }, "human": { "zh-CN": { "name": "stock-analysis", "summary": "Analyze stocks with fundamental and technical analysis. Supports US, China A-shares, and Hong Kong markets. Generate investment reports with key metrics." }, "en": { "name": "stock-analysis", "summary": "Analyze stocks with fundamental and technical analysis. Supports US, China A-shares, and Hong Kong markets. Generate investment reports with key metrics." }, "es": { "name": "stock-analysis", "summary": "Analice acciones con análisis fundamental y técnico. Compatible con los mercados de EE. UU., acciones A de China y Hong Kong. Genere informes de inversión con métricas clave." }, "fr": { "name": "stock-analysis", "summary": "Analysez les actions par analyse fondamentale et technique. Prend en charge les marchés américains, les actions A chinoises et Hong Kong. Génère des rapports d'investissement avec les indicateurs clés." }, "ja": { "name": "stock-analysis", "summary": "ファンダメンタルズ分析とテクニカル分析で株式を分析します。米国市場、中国A株、香港市場に対応し、主要指標を含む投資レポートを生成します。" } }, "whenToUse": "Relevant when working on: finance, us-market, finddata, claude-office-skills.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/us-market/stock-analysis" }, { "id": "skill:last30days", "slug": "skill-last30days", "category": "skill", "marketCategory": null, "tags": [ "sentiment", "research", "finddata", "mvanhorn" ], "canonical": { "brief": "Research what people actually say about any topic in the last 30 days. Pulls posts and engagement from Reddit, X, YouTube, TikTok, Hacker News, Polymarket, GitHub, and the web. Includes a doctor health check to diagnose broken or missing sources." }, "human": { "zh-CN": { "name": "last30days", "summary": "Research what people actually say about any topic in the last 30 days. Pulls posts and engagement from Reddit, X, YouTube, TikTok, Hacker News, Polymarket, GitHub, and the web. Includes a doctor health check to diagnose broken or missing sources." }, "en": { "name": "last30days", "summary": "Research what people actually say about any topic in the last 30 days. Pulls posts and engagement from Reddit, X, YouTube, TikTok, Hacker News, Polymarket, GitHub, and the web. Includes a doctor health check to diagnose broken or missing sources." }, "es": { "name": "last30days", "summary": "Investigue lo que la gente dice realmente sobre cualquier tema en los últimos 30 días. Extrae publicaciones e interacciones de Reddit, X, YouTube, TikTok, Hacker News, Polymarket, GitHub y la web. Incluye un doctor health check para diagnosticar fuentes rotas o ausentes." }, "fr": { "name": "last30days", "summary": "Recherchez ce que les gens disent réellement sur un sujet au cours des 30 derniers jours. Récupère les publications et l'engagement de Reddit, X, YouTube, TikTok, Hacker News, Polymarket, GitHub et du web. Inclut un doctor health check pour diagnostiquer les sources défaillantes ou manquantes." }, "ja": { "name": "last30days", "summary": "過去30日間に人々が実際に何について発言しているかをリサーチします。Reddit、X、YouTube、TikTok、Hacker News、Polymarket、GitHub、ウェブから投稿とエンゲージメントを取得します。壊れたソースや欠落しているソースを診断するためのdoctorヘルスチェックが含まれます。" } }, "whenToUse": "Relevant when working on: sentiment, research, finddata, mvanhorn.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/sentiment/last30days" }, { "id": "skill:data-analysis", "slug": "skill-data-analysis", "category": "skill", "marketCategory": null, "tags": [ "data-analysis", "spreadsheet", "finddata", "claude-office-skills" ], "canonical": { "brief": "Analyze spreadsheet data, generate insights, create visualizations, and build reports from Excel/CSV data." }, "human": { "zh-CN": { "name": "data-analysis", "summary": "Analyze spreadsheet data, generate insights, create visualizations, and build reports from Excel/CSV data." }, "en": { "name": "data-analysis", "summary": "Analyze spreadsheet data, generate insights, create visualizations, and build reports from Excel/CSV data." }, "es": { "name": "data-analysis", "summary": "Analice datos de hojas de cálculo, genere hallazgos, cree visualizaciones y elabore informes a partir de datos Excel/CSV." }, "fr": { "name": "data-analysis", "summary": "Analysez des données de tableurs, générez des enseignements, créez des visualisations et produisez des rapports à partir de données Excel/CSV." }, "ja": { "name": "data-analysis", "summary": "スプレッドシートのデータを分析し、インサイトを生成し、可視化を作成し、Excel/CSVデータからレポートを作成します。" } }, "whenToUse": "Relevant when working on: data-analysis, spreadsheet, finddata, claude-office-skills.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/data-analysis/data-analysis" }, { "id": "skill:exploratory-data-analysis", "slug": "skill-exploratory-data-analysis", "category": "skill", "marketCategory": null, "tags": [ "data-analysis", "eda", "finddata", "k-dense" ], "canonical": { "brief": "Perform bounded, local exploratory analysis of explicitly supported scientific files. Use for redacted CSV/TSV/JSON profiles; optional NumPy, HDF5, FASTA/FASTQ, and basic image metadata inspection; missingness/leakage audits; outlier and transformation sensitivity; and rigorous EDA report scaffolds." }, "human": { "zh-CN": { "name": "exploratory-data-analysis", "summary": "Perform bounded, local exploratory analysis of explicitly supported scientific files. Use for redacted CSV/TSV/JSON profiles; optional NumPy, HDF5, FASTA/FASTQ, and basic image metadata inspection; missingness/leakage audits; outlier and transformation sensitivity; and rigorous EDA report scaffolds. Other domain formats are reference-only and unknown formats fail closed." }, "en": { "name": "exploratory-data-analysis", "summary": "Perform bounded, local exploratory analysis of explicitly supported scientific files. Use for redacted CSV/TSV/JSON profiles; optional NumPy, HDF5, FASTA/FASTQ, and basic image metadata inspection; missingness/leakage audits; outlier and transformation sensitivity; and rigorous EDA report scaffolds. Other domain formats are reference-only and unknown formats fail closed." }, "es": { "name": "exploratory-data-analysis", "summary": "Realice análisis exploratorios acotados y locales de archivos científicos explícitamente compatibles. Úselo para perfiles CSV/TSV/JSON expurgados; inspección opcional de NumPy, HDF5, FASTA/FASTQ y metadatos básicos de imágenes; auditorías de datos faltantes y fugas de información; sensibilidad a atípicos y transformaciones; y andamiajes rigurosos de informes EDA. Los demás formatos de dominio son solo de referencia y los formatos desconocidos fallan de forma cerrada." }, "fr": { "name": "exploratory-data-analysis", "summary": "Effectuez une analyse exploratoire bornée et locale des fichiers scientifiques explicitement pris en charge. À utiliser pour les profils CSV/TSV/JSON expurgés ; l'inspection optionnelle de NumPy, HDF5, FASTA/FASTQ et des métadonnées d'images de base ; les audits de données manquantes et de fuites ; la sensibilité aux valeurs aberrantes et aux transformations ; et des squelettes de rapports EDA rigoureux. Les autres formats de domaine sont fournis à titre de référence uniquement et les formats inconnus échouent en mode fermé." }, "ja": { "name": "exploratory-data-analysis", "summary": "明示的に対応している科学技術ファイルを対象に、範囲を限定したローカルの探索的データ分析を行います。マスキング済みのCSV/TSV/JSONプロファイル、オプションでNumPy・HDF5・FASTA/FASTQ・基本的な画像メタデータの検査、欠損/リークの監査、外れ値や変換への感応度、厳格なEDAレポートのスキャフォールドに使用します。その他のドメイン固有フォーマットは参照のみで、未知のフォーマットはフェイルクローズドします。" } }, "whenToUse": "Relevant when working on: data-analysis, eda, finddata, k-dense.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/data-analysis/exploratory-data-analysis" }, { "id": "skill:statsmodels", "slug": "skill-statsmodels", "category": "skill", "marketCategory": null, "tags": [ "data-analysis", "econometrics", "finddata", "k-dense" ], "canonical": { "brief": "Statistical models library for Python. Use when you need specific model classes (OLS, GLM, mixed models, ARIMA) with detailed diagnostics, residuals, and inference. Best for econometrics, time series, rigorous inference with coefficient tables. For guided statistical test selection with APA reportin" }, "human": { "zh-CN": { "name": "statsmodels", "summary": "Statistical models library for Python. Use when you need specific model classes (OLS, GLM, mixed models, ARIMA) with detailed diagnostics, residuals, and inference. Best for econometrics, time series, rigorous inference with coefficient tables. For guided statistical test selection with APA reporting use statistical-analysis." }, "en": { "name": "statsmodels", "summary": "Statistical models library for Python. Use when you need specific model classes (OLS, GLM, mixed models, ARIMA) with detailed diagnostics, residuals, and inference. Best for econometrics, time series, rigorous inference with coefficient tables. For guided statistical test selection with APA reporting use statistical-analysis." }, "es": { "name": "statsmodels", "summary": "Biblioteca de modelos estadísticos para Python. Úsela cuando necesite clases de modelos específicas (OLS, GLM, modelos mixtos, ARIMA) con diagnósticos detallados, residuos e inferencia. Ideal para econometría, series temporales e inferencia rigurosa con tablas de coeficientes. Para una selección guiada de pruebas estadísticas con informes APA, use statistical-analysis." }, "fr": { "name": "statsmodels", "summary": "Bibliothèque de modèles statistiques pour Python. À utiliser lorsque vous avez besoin de classes de modèles spécifiques (OLS, GLM, modèles mixtes, ARIMA) avec diagnostics détaillés, résidus et inférence. Idéale pour l'économétrie, les séries temporelles et l'inférence rigoureuse avec tableaux de coefficients. Pour une sélection guidée de tests statistiques avec rapport APA, utilisez statistical-analysis." }, "ja": { "name": "statsmodels", "summary": "Python向けの統計モデルライブラリ。詳細な診断、残差、推論を伴う特定のモデルクラス（OLS、GLM、混合モデル、ARIMA）が必要な場合に使用します。計量経済学、時系列分析、係数テーブルを用いた厳密な推論に最適です。APA形式での報告を含む統計検定の選択をガイドしてほしい場合は、statistical-analysisを使用してください。" } }, "whenToUse": "Relevant when working on: data-analysis, econometrics, finddata, k-dense.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/data-analysis/statsmodels" }, { "id": "skill:statistical-analysis", "slug": "skill-statistical-analysis", "category": "skill", "marketCategory": null, "tags": [ "data-analysis", "statistics", "finddata", "k-dense" ], "canonical": { "brief": "Guided statistical analysis for research data - test selection, assumption checking, effect sizes, power analysis, Bayesian alternatives, and APA-formatted reporting. Use whenever a user wants to compare groups, test a hypothesis, analyze experimental or survey data, check statistical assumptions, c" }, "human": { "zh-CN": { "name": "statistical-analysis", "summary": "Guided statistical analysis for research data - test selection, assumption checking, effect sizes, power analysis, Bayesian alternatives, and APA-formatted reporting. Use whenever a user wants to compare groups, test a hypothesis, analyze experimental or survey data, check statistical assumptions, compute required sample sizes, or write up results - even if they never name a specific test. Covers t-tests, ANOVA, chi-square, correlation, regression, non-parametric and Bayesian methods. For low-level model APIs, see the statsmodels and pymc skills." }, "en": { "name": "statistical-analysis", "summary": "Guided statistical analysis for research data - test selection, assumption checking, effect sizes, power analysis, Bayesian alternatives, and APA-formatted reporting. Use whenever a user wants to compare groups, test a hypothesis, analyze experimental or survey data, check statistical assumptions, compute required sample sizes, or write up results - even if they never name a specific test. Covers t-tests, ANOVA, chi-square, correlation, regression, non-parametric and Bayesian methods. For low-level model APIs, see the statsmodels and pymc skills." }, "es": { "name": "statistical-analysis", "summary": "Análisis estadístico guiado para datos de investigación: selección de pruebas, verificación de supuestos, tamaños del efecto, análisis de potencia, alternativas bayesianas e informes en formato APA. Úselo siempre que el usuario quiera comparar grupos, contrastar una hipótesis, analizar datos experimentales o de encuestas, comprobar supuestos estadísticos, calcular tamaños de muestra necesarios o redactar resultados, aunque nunca nombre una prueba específica. Cubre pruebas t, ANOVA, chi-cuadrado, correlación, regresión y métodos no paramétricos y bayesianos. Para API de modelos de bajo nivel, consulte las habilidades statsmodels y pymc." }, "fr": { "name": "statistical-analysis", "summary": "Analyse statistique guidée pour données de recherche : sélection des tests, vérification des hypothèses, tailles d'effet, analyse de puissance, alternatives bayésiennes et rapport au format APA. À utiliser chaque fois que l'utilisateur veut comparer des groupes, tester une hypothèse, analyser des données expérimentales ou d'enquête, vérifier des hypothèses statistiques, calculer des tailles d'échantillon requises ou rédiger des résultats, même s'il ne nomme jamais un test précis. Couvre les tests t, l'ANOVA, le khi-deux, la corrélation, la régression et les méthodes non paramétriques et bayésiennes. Pour les API de modèles de bas niveau, voir les compétences statsmodels et pymc." }, "ja": { "name": "statistical-analysis", "summary": "リサーチデータのためのガイド付き統計分析：検定の選択、前提条件の確認、効果量、検出力分析、ベイズ的代替手法、APA形式の報告をカバーします。ユーザーがグループの比較、仮説の検定、実験データや調査データの分析、統計的前提の確認、必要なサンプルサイズの計算、結果のまとめを望む場合に、特定の検定名が挙げられなくても使用します。t検定、ANOVA、カイ二乗検定、相関、回帰、ノンパラメトリック手法、ベイズ手法をカバーします。低レベルのモデルAPIについては、statsmodelsおよびpymcスキルを参照してください。" } }, "whenToUse": "Relevant when working on: data-analysis, statistics, finddata, k-dense.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/data-analysis/statistical-analysis" }, { "id": "skill:legal-harness-init", "slug": "skill-legal-harness-init", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Initialize or incrementally govern AGENTS.md/CLAUDE.md for legal practitioners: identify the current AI harness, distinguish user-level, project-level, and team-level instructions, generate a minimal legal safety baseline, safely merge managed blocks, and verify permissions, confidentiality, informa" }, "human": { "zh-CN": { "name": "legal-harness-init", "summary": "面向法律工作者初始化或增量治理 AGENTS.md/CLAUDE.md：识别当前 AI harness，区分用户级、项目级和团队级指令，生成最小法律安全基线，安全合并受管区块，并在新会话中验证权限、保密、信息缺口和回溯行为。用户明确提到法律工作且要配置 harness、AGENTS.md、CLAUDE.md、agent 协作规则或 AI 使用基线时使用。不要用于合同审查、案件分析、文书起草、项目脚手架或 Skill 开发。" }, "en": { "name": "legal-harness-init", "summary": "Initialize or incrementally govern AGENTS.md/CLAUDE.md for legal practitioners: identify the current AI harness, distinguish user-level, project-level, and team-level instructions, generate a minimal legal safety baseline, safely merge managed blocks, and verify permissions, confidentiality, information gaps, and trace-back behavior in a new session. Use when the user explicitly mentions legal work and wants to configure a harness, AGENTS.md, CLAUDE.md, agent collaboration rules, or an AI usage baseline. Do not use for contract review, case analysis, document drafting, project scaffolding, or skill development." }, "es": { "name": "legal-harness-init", "summary": "Inicialice o gobierne de forma incremental AGENTS.md/CLAUDE.md para profesionales del derecho: identifique el harness de IA actual, distinga instrucciones a nivel de usuario, de proyecto y de equipo, genere una línea base mínima de seguridad jurídica, fusione de forma segura los bloques gestionados y verifique en una nueva sesión permisos, confidencialidad, lagunas de información y comportamiento de trazabilidad. Úselo cuando el usuario mencione explícitamente trabajo legal y quiera configurar un harness, AGENTS.md, CLAUDE.md, reglas de colaboración de agentes o una línea base de uso de IA. No lo use para revisión de contratos, análisis de casos, redacción de documentos, andamiaje de proyectos ni desarrollo de habilidades." }, "fr": { "name": "legal-harness-init", "summary": "Initialisez ou gouvernez de manière incrémentale AGENTS.md/CLAUDE.md pour les professionnels du droit : identifiez le harness IA actuel, distinguez les instructions au niveau utilisateur, projet et équipe, générez une base minimale de sécurité juridique, fusionnez en toute sécurité les blocs gérés et vérifiez dans une nouvelle session les permissions, la confidentialité, les lacunes d'information et le comportement de traçabilité. À utiliser lorsque l'utilisateur mentionne explicitement un travail juridique et souhaite configurer un harness, AGENTS.md, CLAUDE.md, des règles de collaboration d'agents ou une base d'utilisation de l'IA. Ne pas utiliser pour la révision de contrats, l'analyse de dossiers, la rédaction d'actes, l'échafaudage de projets ou le développement de compétences." }, "ja": { "name": "legal-harness-init", "summary": "法律実務者向けにAGENTS.md/CLAUDE.mdを初期化または段階的にガバナンスします：現在のAIハーネスを識別し、ユーザーレベル・プロジェクトレベル・チームレベルの指示を区別し、法的安全性の最小ベースラインを生成し、管理対象ブロックを安全にマージし、新しいセッションで権限・機密保持・情報の欠落・遡及動作を検証します。ユーザーが法律業務に触れた上で、ハーネス、AGENTS.md、CLAUDE.md、エージェント協作ルール、AI利用ベースラインの設定を明示的に求める場合に使用します。契約審査、案件分析、文書の起草、プロジェクトのスキャフォールド、スキル開発には使用しないでください。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:legal-ocr", "slug": "skill-legal-ocr", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Use this skill when the user needs OCR, scan recognition, text recognition from images, document recognition, or conversion of PDFs, images, Office documents, or URLs into Markdown. When legal materials are detected, conservative optimization of legal terminology and document structure may be applie" }, "human": { "zh-CN": { "name": "legal-ocr", "summary": "本技能应在用户需要 OCR、扫描识别、图片文字识别、文档识别，或将 PDF、图片、Office 文档、URL 转换为 Markdown 时使用。检测到法律材料时可进行保守的法律术语与文书结构优化。不要用于法律事实判断、补写缺失内容、语义改写、印章深度识别或图表实体分析。" }, "en": { "name": "legal-ocr", "summary": "Use this skill when the user needs OCR, scan recognition, text recognition from images, document recognition, or conversion of PDFs, images, Office documents, or URLs into Markdown. When legal materials are detected, conservative optimization of legal terminology and document structure may be applied. Do not use for legal fact-finding, filling in missing content, semantic rewriting, deep seal recognition, or entity analysis of charts." }, "es": { "name": "legal-ocr", "summary": "Use esta habilidad cuando el usuario necesite OCR, reconocimiento de escaneos, reconocimiento de texto en imágenes, reconocimiento de documentos o conversión de PDF, imágenes, documentos de Office o URL a Markdown. Cuando se detecten materiales legales, puede aplicarse una optimización conservadora de la terminología jurídica y de la estructura documental. No la use para juicios de hechos jurídicos, completar contenido faltante, reescritura semántica, reconocimiento profundo de sellos ni análisis de entidades en gráficos." }, "fr": { "name": "legal-ocr", "summary": "Utilisez cette compétence lorsque l'utilisateur a besoin d'OCR, de reconnaissance de scans, de reconnaissance de texte dans des images, de reconnaissance de documents ou de conversion de PDF, d'images, de documents Office ou d'URL en Markdown. En présence de documents juridiques détectés, une optimisation conservative de la terminologie juridique et de la structure documentaire peut être appliquée. Ne pas l'utiliser pour l'appréciation de faits juridiques, le remplissage de contenu manquant, la réécriture sémantique, la reconnaissance approfondie de cachets ni l'analyse d'entités dans des graphiques." }, "ja": { "name": "legal-ocr", "summary": "ユーザーがOCR、スキャン認識、画像内の文字認識、ドキュメント認識を必要とする場合や、PDF・画像・Office文書・URLをMarkdownに変換したい場合に使用するスキルです。法的資料を検出した場合は、法律用語と文書構造の保守的な最適化を行うことができます。法律上の事実認定、欠落内容の補完、意味的な書き換え、印影の詳細認識、図表のエンティティ分析には使用しないでください。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:pdf-organizer", "slug": "skill-pdf-organizer", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Use when legal PDFs need organizing: detects the text layer, generates page indexes, organization drafts, and downstream handoff documents; splits, merges, or renames post-OCR dual-layer scans by content with standardized naming; can perform rotation and skew correction, but does not do OCR or compr" }, "human": { "zh-CN": { "name": "pdf-organizer", "summary": "当需要整理法律 PDF 时使用：检测文字层，生成页面索引、整理草稿和下游交接文件，按内容拆分、合并或直接重命名 OCR 后双层扫描件并规范命名；可做旋转与倾斜校正，不做 OCR 或压缩。" }, "en": { "name": "pdf-organizer", "summary": "Use when legal PDFs need organizing: detects the text layer, generates page indexes, organization drafts, and downstream handoff documents; splits, merges, or renames post-OCR dual-layer scans by content with standardized naming; can perform rotation and skew correction, but does not do OCR or compression." }, "es": { "name": "pdf-organizer", "summary": "Úsela cuando haya que organizar PDF legales: detecta la capa de texto, genera índices de páginas, borradores de organización y documentos de entrega para fases posteriores; divide, fusiona o renombra directamente escaneos duales post-OCR por contenido con nomenclatura normalizada; puede realizar corrección de rotación e inclinación, pero no hace OCR ni compresión." }, "fr": { "name": "pdf-organizer", "summary": "À utiliser pour organiser des PDF juridiques : détecte la couche de texte, génère des index de pages, des brouillons d'organisation et des documents de transmission aux étapes en aval ; découpe, fusionne ou renomme directement les scans bicouches post-OCR par contenu avec une nomenclature normalisée ; peut effectuer des corrections de rotation et d'inclinaison, mais ne fait ni OCR ni compression." }, "ja": { "name": "pdf-organizer", "summary": "法律PDFを整理する必要があるときに使用します：文字レイヤーを検出し、ページ索引、整理ドラフト、下流工程への引き継ぎ資料を生成し、OCR済み二層スキャンを内容に基づいて分割・結合・リネームし、命名規則を統一します。回転補正や傾き補正は可能ですが、OCRや圧縮は行いません。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:china-legal-skills", "slug": "skill-china-legal-skills", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Chinese legal workflow skills for lawyers, legal counsel, litigation, criminal defense, labor disputes, bankruptcy, contract review, compliance, legal research, and legal document drafting." }, "human": { "zh-CN": { "name": "china-legal-skills", "summary": "Chinese legal workflow skills for lawyers, legal counsel, litigation, criminal defense, labor disputes, bankruptcy, contract review, compliance, legal research, and legal document drafting." }, "en": { "name": "china-legal-skills", "summary": "Chinese legal workflow skills for lawyers, legal counsel, litigation, criminal defense, labor disputes, bankruptcy, contract review, compliance, legal research, and legal document drafting." }, "es": { "name": "china-legal-skills", "summary": "Habilidades de flujos de trabajo jurídicos chinos para abogados, asesores legales, litigios, defensa penal, disputas laborales, concursos de acreedores, revisión de contratos, cumplimiento, investigación jurídica y redacción de documentos legales." }, "fr": { "name": "china-legal-skills", "summary": "Compétences de workflows juridiques chinois pour avocats, conseils juridiques, contentieux, défense pénale, litiges du travail, faillite, révision de contrats, conformité, recherche juridique et rédaction de documents juridiques." }, "ja": { "name": "china-legal-skills", "summary": "弁護士・法務担当者向けの中国法務ワークフロースキル集。訴訟、刑事弁護、労働紛争、破産、契約審査、コンプライアンス、リーガルリサーチ、法務文書の作成をカバーします。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:product-legal-compliance", "slug": "skill-product-legal-compliance", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Product legal and compliance skill for the Chinese market. Used for product launches, feature compliance, default-on settings, customer logo/case-study display, review of PRDs/page flows/user agreements/privacy policies/help centers, compliance review of product materials, in-depth assessment of pro" }, "human": { "zh-CN": { "name": "product-legal-compliance", "summary": "中国境内产品法务与产品合规 Skill。用于产品上线、功能合规、默认开启、客户 Logo/客户案例展示、PRD/页面流程/用户协议/隐私政策/帮助中心审查、产品材料合规审查、高合规行业产品深度评估、营销文案产品场景初筛等任务。适用于用户询问“这个功能能不能做”“能不能把客户 logo 放官网”“产品材料是否合法合规”“医疗/金融/教育/AI/平台类产品是否能上线”等问题。" }, "en": { "name": "product-legal-compliance", "summary": "Product legal and compliance skill for the Chinese market. Used for product launches, feature compliance, default-on settings, customer logo/case-study display, review of PRDs/page flows/user agreements/privacy policies/help centers, compliance review of product materials, in-depth assessment of products in high-compliance industries, and initial screening of product scenarios in marketing copy. Suited to questions such as \"can this feature be built\", \"can we put a customer's logo on the official website\", \"are the product materials lawful and compliant\", and \"can a healthcare/finance/education/AI/platform product go live\"." }, "es": { "name": "product-legal-compliance", "summary": "Habilidad de asesoría jurídica y cumplimiento de producto para el mercado de China. Se usa para lanzamientos de producto, cumplimiento de funcionalidades, ajustes activados por defecto, exhibición de logos/casos de clientes, revisión de PRD/flujos de página/acuerdos de usuario/políticas de privacidad/centros de ayuda, revisión de cumplimiento de materiales de producto, evaluación en profundidad de productos de sectores de alta exigencia normativa y preselección de escenarios de producto en textos de marketing. Adecuada para preguntas como «¿se puede hacer esta función?», «¿podemos poner el logo de un cliente en la web oficial?», «¿los materiales del producto son legales y conformes?» o «¿puede lanzarse un producto de salud/finanzas/educación/IA/plataforma?»." }, "fr": { "name": "product-legal-compliance", "summary": "Compétence de juridique et conformité produit pour le marché chinois. Utilisée pour les lancements de produit, la conformité des fonctionnalités, les réglages activés par défaut, l'affichage de logos/études de cas clients, la revue des PRD/parcours de pages/conditions d'utilisation/politiques de confidentialité/centres d'aide, la revue de conformité des supports produit, l'évaluation approfondie des produits de secteurs fortement réglementés et la présélection des scénarios produit dans les textes marketing. Convient aux questions telles que « peut-on développer cette fonctionnalité », « peut-on afficher le logo d'un client sur le site officiel », « les supports produit sont-ils licites et conformes », « un produit santé/finance/éducation/IA/plateforme peut-il être lancé »." }, "ja": { "name": "product-legal-compliance", "summary": "中国国内向けのプロダクト法務・コンプライアンススキル。製品リリース、機能のコンプライアンス、デフォルトON設定、顧客ロゴ/導入事例の掲載、PRD/ページフロー/利用規約/プライバシーポリシー/ヘルプセンターの審査、製品資材のコンプライアンス審査、高コンプライアンス業界製品の詳細評価、マーケティングコピーにおける製品シナリオの一次スクリーニングなどのタスクに使用します。「この機能は作れるか」「顧客ロゴを公式サイトに載せられるか」「製品資材は合法・コンプライアンスに適合しているか」「医療/金融/教育/AI/プラットフォーム製品はリリースできるか」といった質問に適しています。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:investigation-stage-criminal-defense", "slug": "skill-investigation-stage-criminal-defense", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Criminal defense skill for the investigation stage, covering the full client-meeting workflow, legal analysis of the case and sentencing prediction, management of arrest-approval timing milestones, legal document generation, and practice risk prevention — a complete set of defense work standards." }, "human": { "zh-CN": { "name": "investigation-stage-criminal-defense", "summary": "刑事辩护侦查阶段工作Skill，涵盖会见全流程、案件法律分析与量刑预判、批捕时间节点管理、法律文书生成、执业风险防范等完整辩护工作规范" }, "en": { "name": "investigation-stage-criminal-defense", "summary": "Criminal defense skill for the investigation stage, covering the full client-meeting workflow, legal analysis of the case and sentencing prediction, management of arrest-approval timing milestones, legal document generation, and practice risk prevention — a complete set of defense work standards." }, "es": { "name": "investigation-stage-criminal-defense", "summary": "Habilidad de defensa penal para la etapa de investigación, que abarca el flujo completo de entrevistas con el cliente, el análisis jurídico del caso y la predicción de pena, la gestión de los hitos temporales de aprobación de arresto, la generación de documentos legales y la prevención de riesgos profesionales: un conjunto completo de normas de trabajo de la defensa." }, "fr": { "name": "investigation-stage-criminal-defense", "summary": "Compétence de défense pénale pour la phase d'enquête, couvrant le processus complet des entretiens avec le client, l'analyse juridique du dossier et la prédiction de la peine, la gestion des jalons temporels d'approbation d'arrestation, la génération de documents juridiques et la prévention des risques professionnels : un ensemble complet de normes de travail de la défense." }, "ja": { "name": "investigation-stage-criminal-defense", "summary": "刑事弁護の捜査段階に対応するスキル。接見の全フロー、案件の法的分析と量刑予測、逮捕認可のタイムライン管理、法律文書の生成、実務リスク防止など、弁護活動の完全な作業規範をカバーします。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:preliminary-legal-analysis", "slug": "skill-preliminary-legal-analysis", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Performs systematic legal analysis based on the fused framework of the \"six-source system\", the 6-step claim-basis method, and the nine-step element-based adjudication method; links to the master controller's current-matter records and retrospective ledger to generate a legal service proposal as a W" }, "human": { "zh-CN": { "name": "preliminary-legal-analysis", "summary": "基于\"六来源体系\"、请求权基础6步法与要件审判九步法融合框架，进行系统性法律分析，按总控当前事项记录和复盘台账衔接，生成法律服务建议书Word文档。当用户发送客户编号并提及\"初步法律分析\"、\"法律分析\"、\"请求权分析\"等指令时触发使用。" }, "en": { "name": "preliminary-legal-analysis", "summary": "Performs systematic legal analysis based on the fused framework of the \"six-source system\", the 6-step claim-basis method, and the nine-step element-based adjudication method; links to the master controller's current-matter records and retrospective ledger to generate a legal service proposal as a Word document. Triggers when the user sends a client number and mentions instructions such as \"preliminary legal analysis\", \"legal analysis\", or \"claim analysis\"." }, "es": { "name": "preliminary-legal-analysis", "summary": "Realiza un análisis jurídico sistemático basado en el marco fusionado del «sistema de seis fuentes», el método en 6 pasos de la base de la pretensión y el método en nueve pasos del enjuiciamiento por elementos; se enlaza con los registros de asuntos actuales del controlador maestro y el libro de retrospectiva para generar una propuesta de servicios jurídicos como documento Word. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como «análisis jurídico preliminar», «análisis jurídico» o «análisis de la pretensión»." }, "fr": { "name": "preliminary-legal-analysis", "summary": "Réalise une analyse juridique systématique fondée sur le cadre fusionné du « système des six sources », la méthode en 6 étapes du fondement de la prétention et la méthode en neuf étapes du jugement par éléments ; se relie aux enregistrements des affaires en cours du contrôleur maître et au registre de rétrospection pour générer une proposition de services juridiques sous forme de document Word. Se déclenche lorsque l'utilisateur envoie un numéro de client et mentionne des instructions telles que « analyse juridique préliminaire », « analyse juridique » ou « analyse de la prétention »." }, "ja": { "name": "preliminary-legal-analysis", "summary": "「六来源体系」、請求権基礎6ステップ法、要件審理9ステップ法を融合したフレームワークに基づき、体系的な法的分析を行い、マスターコントロールの現在事項記録と振り返り台帳に連携して、リーガルサービス提案書をWord文書として生成します。ユーザーが顧客番号を送り、「初期法律分析」「法律分析」「請求権分析」などの指示に言及したときに起動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:labor-arbitration-case-management", "slug": "skill-labor-arbitration-case-management", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Full-workflow management skill for labor dispute arbitration, covering mandatory arbitration-prerequisite detection, arbitration limitation-period calculation, automatic one-award-finality determination, generation of arbitration applications/responses (dual W/E roles), arbitration claim checklists," }, "human": { "zh-CN": { "name": "labor-arbitration-case-management", "summary": "劳动争议仲裁程序全流程管理Skill，覆盖仲裁前置强制检测、仲裁时效计算、一裁终局自动判断、仲裁申请书/答辩书生成（W/E双角色）、仲裁请求清单、仲裁→诉讼衔接路由（L9五分支）、仲裁策略报告和用人单位仲裁前风险敞口评估。当用户发送客户编号并提及\"仲裁申请\"、\"仲裁答辩\"、\"仲裁策略\"、\"仲裁时效\"、\"公司仲裁风险\"、\"风险敞口\"等指令时触发使用。" }, "en": { "name": "labor-arbitration-case-management", "summary": "Full-workflow management skill for labor dispute arbitration, covering mandatory arbitration-prerequisite detection, arbitration limitation-period calculation, automatic one-award-finality determination, generation of arbitration applications/responses (dual W/E roles), arbitration claim checklists, arbitration-to-litigation routing (L9 five branches), arbitration strategy reports, and pre-arbitration risk-exposure assessment for employers. Triggers when the user sends a client number and mentions instructions such as \"arbitration application\", \"arbitration response\", \"arbitration strategy\", \"arbitration limitation period\", \"company arbitration risk\", or \"risk exposure\"." }, "es": { "name": "labor-arbitration-case-management", "summary": "Habilidad de gestión de todo el flujo del procedimiento arbitral en disputas laborales: cubre la detección obligatoria del requisito previo de arbitraje, el cálculo de los plazos de prescripción arbitral, la determinación automática de la firmeza del laudo (un solo arbitraje con firmeza), la generación de solicitudes/contestaciones arbitrales (doble rol W/E), listas de pretensiones arbitrales, el enrutamiento arbitraje→litigio (L9, cinco ramas), informes de estrategia arbitral y la evaluación de la exposición al riesgo prearbitral del empleador. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como «solicitud de arbitraje», «contestación arbitral», «estrategia arbitral», «plazo de prescripción arbitral», «riesgo arbitral de la empresa» o «exposición al riesgo»." }, "fr": { "name": "labor-arbitration-case-management", "summary": "Compétence de gestion du flux complet de la procédure d'arbitrage des litiges du travail : couvre la détection obligatoire du préalable arbitral, le calcul des délais de prescription de l'arbitrage, la détermination automatique du caractère définitif de la sentence (un seul arbitrage définitif), la génération de demandes/mémoires en défense arbitraux (double rôle W/E), les listes de demandes arbitrales, le routage arbitrage→contentieux (L9, cinq branches), les rapports de stratégie d'arbitrage et l'évaluation de l'exposition au risque pré-arbitrage de l'employeur. Se déclenche lorsque l'utilisateur envoie un numéro de client et mentionne des instructions telles que « demande d'arbitrage », « défense arbitrale », « stratégie d'arbitrage », « délai de prescription de l'arbitrage », « risque d'arbitrage de l'entreprise » ou « exposition au risque »." }, "ja": { "name": "labor-arbitration-case-management", "summary": "労働紛争仲裁手続きの全フロー管理スキル。仲裁前置の強制チェック、仲裁時効の計算、一裁終局の自動判定、仲裁申立書/答弁書の生成（W/E二役）、仲裁請求リスト、仲裁から訴訟への連携ルーティング（L9五分岐）、仲裁戦略レポート、使用者側の仲裁前リスクエクスポージャー評価をカバーします。ユーザーが顧客番号を送り、「仲裁申立て」「仲裁答弁」「仲裁戦略」「仲裁時効」「会社の仲裁リスク」「リスクエクスポージャー」などの指示に言及したときに起動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:labor-dispute-evidence-system", "slug": "skill-labor-dispute-evidence-system", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Dedicated evidence-management skill for labor dispute cases, covering burden-of-proof reversal rules, automatic generation of evidence collection checklists by 12 dispute types, three-tier assessment of evidentiary weight, automatic detection of four types of special protection, the three elements o" }, "human": { "zh-CN": { "name": "labor-dispute-evidence-system", "summary": "劳动争议案件证据管理专属Skill，覆盖举证责任倒置规则、按12类争议类型自动生成证据收集清单、证据效力三级评估、四类特殊保护自动检测、规章制度合法性审查三要件、劳动者代理证据完整性检查和用人单位争议/合规证据清单。输出举证责任分配表、证据收集清单、特殊保护检测报告、公司端事实与制度证据缺口表。当用户发送客户编号并提及\"举证责任\"、\"劳动争议证据\"、\"证据清单\"、\"特殊保护\"、\"员工手册证据\"、\"公司端证据\"等指令时触发使用。" }, "en": { "name": "labor-dispute-evidence-system", "summary": "Dedicated evidence-management skill for labor dispute cases, covering burden-of-proof reversal rules, automatic generation of evidence collection checklists by 12 dispute types, three-tier assessment of evidentiary weight, automatic detection of four types of special protection, the three elements of legality review of internal rules, evidence completeness checks for worker representation, and employer dispute/compliance evidence checklists. Outputs a burden-of-proof allocation table, evidence collection checklists, special-protection detection reports, and company-side fact-and-policy evidence gap tables. Triggers when the user sends a client number and mentions instructions such as \"burden of proof\", \"labor dispute evidence\", \"evidence checklist\", \"special protection\", \"employee handbook evidence\", or \"company-side evidence\"." }, "es": { "name": "labor-dispute-evidence-system", "summary": "Habilidad dedicada a la gestión de pruebas en disputas laborales: cubre las reglas de inversión de la carga de la prueba, la generación automática de listas de recolección de pruebas según 12 tipos de disputa, la evaluación en tres niveles de la eficacia probatoria, la detección automática de cuatro tipos de protección especial, los tres elementos de la revisión de legalidad de los reglamentos internos, la verificación de integridad de pruebas para la representación del trabajador y las listas de pruebas de disputa/cumplimiento del empleador. Genera una tabla de asignación de la carga de la prueba, listas de recolección de pruebas, informes de detección de protección especial y tablas de brechas de pruebas fácticas y normativas del lado empresarial. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como «carga de la prueba», «pruebas de la disputa laboral», «lista de pruebas», «protección especial», «pruebas del manual del empleado» o «pruebas del lado de la empresa»." }, "fr": { "name": "labor-dispute-evidence-system", "summary": "Compétence dédiée à la gestion des preuves dans les litiges du travail : couvre les règles d'inversion de la charge de la preuve, la génération automatique de listes de collecte de preuves selon 12 types de litiges, l'évaluation à trois niveaux de la force probante, la détection automatique de quatre types de protection spéciale, les trois éléments de l'examen de légalité du règlement intérieur, le contrôle d'intégrité des preuves pour la représentation du travailleur et les listes de preuves de litige/conformité de l'employeur. Produit un tableau d'imputation de la charge de la preuve, des listes de collecte de preuves, des rapports de détection de protection spéciale et des tableaux de lacunes probatoires factuelles et réglementaires côté entreprise. Se déclenche lorsque l'utilisateur envoie un numéro de client et mentionne des instructions telles que « charge de la preuve », « preuves du litige du travail », « liste de preuves », « protection spéciale », « preuves du manuel du personnel » ou « preuves côté entreprise »." }, "ja": { "name": "labor-dispute-evidence-system", "summary": "労働紛争案件の証拠管理に特化したスキル。挙証責任の転換ルール、12類型の紛争タイプ別の証拠収集リスト自動生成、証拠証明力の3段階評価、4種類の特別保護の自動検出、就業規則の合法性審査3要件、労働者代理側の証拠完全性チェック、使用者側の紛争/コンプライアンス証拠リストをカバーします。挙証責任配分表、証拠収集リスト、特別保護検出レポート、会社側の事実・制度証拠ギャップ表を出力します。ユーザーが顧客番号を送り、「挙証責任」「労働紛争の証拠」「証拠リスト」「特別保護」「従業員ハンドブックの証拠」「会社側の証拠」などの指示に言及したときに起動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:labor-dispute-litigation", "slug": "skill-labor-dispute-litigation", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Professional workflow-orchestration skill for labor dispute litigation. Integrates 17 sub-skills: reuses 14 existing general skills plus 3 labor-dispute-specific skills (labor dispute arbitration procedure management, labor relation determination and economic compensation calculation, and the labor " }, "human": { "zh-CN": { "name": "labor-dispute-litigation", "summary": "劳动争议诉讼专业流程编排 Skill。整合17个子Skill：复用14个现有通用Skill + 3个劳动争议专属Skill（劳动争议仲裁程序管理、劳动关系认定与经济补偿计算、劳动争议证据体系）。用于判断劳动争议阶段（协商→调解→仲裁→一审→二审，L0-L15/LM系列）、识别W/E代理角色、区分劳动者代理、用人单位争议代理和用人单位日常劳动合规任务、推荐下一步子Skill、提示关键节点确认和期限；事项建档、路径、文件读取、来源披露、OCR 校正和缺口归档由「法律工作总控」统一处理。" }, "en": { "name": "labor-dispute-litigation", "summary": "Professional workflow-orchestration skill for labor dispute litigation. Integrates 17 sub-skills: reuses 14 existing general skills plus 3 labor-dispute-specific skills (labor dispute arbitration procedure management, labor relation determination and economic compensation calculation, and the labor dispute evidence system). Used to determine the labor dispute stage (negotiation → mediation → arbitration → first instance → second instance, L0-L15/LM series), identify W/E representation roles, distinguish worker representation, employer dispute representation, and employer routine labor compliance tasks, recommend the next sub-skill, and flag key-node confirmations and deadlines; matter filing, paths, file reading, source disclosure, OCR correction, and gap archiving are handled uniformly by the \"legal work master controller\"." }, "es": { "name": "labor-dispute-litigation", "summary": "Habilidad de orquestación de flujo profesional para el litigio de disputas laborales. Integra 17 subhabilidades: reutiliza 14 habilidades generales existentes más 3 habilidades específicas de disputas laborales (gestión del procedimiento arbitral de disputas laborales, determinación de la relación laboral y cálculo de la compensación económica, y sistema de pruebas de disputas laborales). Se usa para determinar la etapa de la disputa laboral (negociación → mediación → arbitraje → primera instancia → segunda instancia, serie L0-L15/LM), identificar los roles de representación W/E, distinguir la representación del trabajador, la representación en disputas del empleador y las tareas de cumplimiento laboral cotidiano del empleador, recomendar la siguiente subhabilidad y señalar confirmaciones de nodos clave y plazos; el expediente del asunto, las rutas, la lectura de archivos, la divulgación de fuentes, la corrección OCR y el archivo de brechas se gestionan de forma unificada por el «controlador maestro de trabajo legal»." }, "fr": { "name": "labor-dispute-litigation", "summary": "Compétence d'orchestration professionnelle de flux pour le contentieux des litiges du travail. Intègre 17 sous-compétences : réutilise 14 compétences générales existantes plus 3 compétences propres aux litiges du travail (gestion de la procédure d'arbitrage des litiges du travail, détermination de la relation de travail et calcul de l'indemnité économique, et système de preuves des litiges du travail). Sert à déterminer la phase du litige du travail (négociation → médiation → arbitrage → première instance → appel, série L0-L15/LM), identifier les rôles de représentation W/E, distinguer représentation du travailleur, représentation contentieuse de l'employeur et tâches de conformité laborale quotidienne de l'employeur, recommander la sous-compétence suivante et signaler les confirmations de jalons clés et les échéances ; l'ouverture du dossier, les chemins, la lecture de fichiers, la divulgation des sources, la correction OCR et l'archivage des lacunes sont gérés uniformément par le « contrôleur maître du travail juridique »." }, "ja": { "name": "labor-dispute-litigation", "summary": "労働紛争訴訟のためのプロフェッショナルなワークフローオーケストレーションスキル。17のサブスキルを統合します：既存の汎用スキル14個＋労働紛争専用スキル3個（労働紛争仲裁手続き管理、労働関係の認定と経済補償の計算、労働紛争証拠体系）を再利用します。労働紛争の段階判定（協商→調停→仲裁→一審→二審、L0-L15/LMシリーズ）、W/E代理役割の識別、労働者代理・使用者側紛争代理・使用者側の日常的労働コンプライアンス業務の区別、次のサブスキルの推奨、重要ノードの確認と期限の提示に使用します。案件のファイル作成、パス、ファイル読み取り、ソースの開示、OCR補正、ギャップのアーカイブは「法律業務マスターコントロール」が一元的に処理します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:labor-relation-severance-calculator", "slug": "skill-labor-relation-severance-calculator", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Core analysis tool for labor dispute cases, covering three-factor subordination analysis of the labor relation, automatic classification into 12 dispute types, working-hours system characterization, wage-and-hours Q&A, an economic compensation calculation engine (N/N+1/2N determination + monthly wag" }, "human": { "zh-CN": { "name": "labor-relation-severance-calculator", "summary": "劳动争议案件核心分析工具，覆盖劳动关系三性从属分析、12类争议自动分类、工时制度定性、工资工时问答、经济补偿计算引擎（N/N+1/2N判断+月工资基数+三层封顶）、加班费三档计算、双倍工资计算。当用户发送客户编号并提及\"劳动关系认定\"、\"经济补偿\"、\"补偿计算\"、\"加班费计算\"、\"工资工时\"、\"发薪日\"、\"特殊工时\"等指令时触发使用。" }, "en": { "name": "labor-relation-severance-calculator", "summary": "Core analysis tool for labor dispute cases, covering three-factor subordination analysis of the labor relation, automatic classification into 12 dispute types, working-hours system characterization, wage-and-hours Q&A, an economic compensation calculation engine (N/N+1/2N determination + monthly wage base + three-tier caps), three-tier overtime pay calculation, and double-wage calculation. Triggers when the user sends a client number and mentions instructions such as \"labor relation determination\", \"economic compensation\", \"compensation calculation\", \"overtime pay calculation\", \"wages and working hours\", \"payday\", or \"special working hours\"." }, "es": { "name": "labor-relation-severance-calculator", "summary": "Herramienta central de análisis para casos de disputas laborales: cubre el análisis de subordinación de tres elementos de la relación laboral, la clasificación automática en 12 tipos de disputa, la caracterización del régimen de jornada, preguntas y respuestas sobre salarios y jornadas, un motor de cálculo de la compensación económica (determinación N/N+1/2N + base salarial mensual + tres niveles de tope), el cálculo de horas extra en tres tramos y el cálculo del doble salario. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como «determinación de la relación laboral», «compensación económica», «cálculo de la compensación», «cálculo de horas extra», «salarios y jornadas», «día de pago» o «jornadas especiales»." }, "fr": { "name": "labor-relation-severance-calculator", "summary": "Outil d'analyse central pour les dossiers de litiges du travail : couvre l'analyse de subordination à trois critères de la relation de travail, la classification automatique en 12 types de litiges, la caractérisation du régime d'horaire, les questions-réponses salaires et horaires, un moteur de calcul de l'indemnité économique (détermination N/N+1/2N + base salariale mensuelle + plafonds à trois niveaux), le calcul des heures supplémentaires à trois paliers et le calcul du salaire double. Se déclenche lorsque l'utilisateur envoie un numéro de client et mentionne des instructions telles que « détermination de la relation de travail », « indemnité économique », « calcul de l'indemnité », « calcul des heures supplémentaires », « salaires et horaires », « jour de paie » ou « horaires spéciaux »." }, "ja": { "name": "labor-relation-severance-calculator", "summary": "労働紛争案件のコア分析ツール。労働関係の三性（従属性）分析、12類型の紛争自動分類、労働時間制度の定性、賃金・労働時間のQ&A、経済補償計算エンジン（N/N+1/2Nの判定＋月給ベース＋3段階の上限）、残業代の3区分計算、二倍賃金の計算をカバーします。ユーザーが顧客番号を送り、「労働関係の認定」「経済補償」「補償計算」「残業代計算」「賃金・労働時間」「給料日」「特殊労働時間」などの指示に言及したときに起動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:pa1nrui1-contract-review", "slug": "skill-pa1nrui1-contract-review", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Performs in-depth contract review based on Zhang Haiyan's four-step review method plus the He Li/Chang Jinguang three-perspective analysis method. Supports company verification, automatic multilingual detection, Mermaid business flow diagrams, two-stage review opinions, contract overviews, client-re" }, "human": { "zh-CN": { "name": "pa1nrui1-contract-review", "summary": "基于张海燕四步审查法+何力/常金光三观分析法，对合同进行深度审查。支持企业核验、多语言自动检测、Mermaid业务流程图、二段式审核意见、合同概览、客户可读合同编号、业务摘要、22类合同专项审查、续约提醒和 Word 修订模式审核红线稿。飞书正式交付物必须包含问题卡片列表和业务流程图；用户选择修订模式时，以原合同为只读来源，先生成 redline-plan.json，再运行 scripts/redline/apply_redline_plan.py 生成带真实修订痕迹和批注的 Word 审核修订稿，并完成红线 QA。其他记录保存为本地 Markdown。" }, "en": { "name": "pa1nrui1-contract-review", "summary": "Performs in-depth contract review based on Zhang Haiyan's four-step review method plus the He Li/Chang Jinguang three-perspective analysis method. Supports company verification, automatic multilingual detection, Mermaid business flow diagrams, two-stage review opinions, contract overviews, client-readable contract numbers, business summaries, specialized review for 22 contract types, renewal reminders, and Word tracked-changes redline review drafts. Feishu formal deliverables must include the issue card list and the business flow diagram; when the user selects revision mode, the original contract is treated as a read-only source, redline-plan.json is generated first, then scripts/redline/apply_redline_plan.py is run to produce a Word review revision with real tracked changes and comments, completing redline QA. Other records are saved as local Markdown." }, "es": { "name": "pa1nrui1-contract-review", "summary": "Realiza una revisión en profundidad de contratos basada en el método de revisión en cuatro pasos de Zhang Haiyan más el método de análisis de tres perspectivas de He Li/Chang Jinguang. Admite verificación empresarial, detección automática multilingüe, diagramas de flujo de negocio Mermaid, dictámenes de revisión en dos etapas, resúmenes del contrato, números de contrato legibles para el cliente, resúmenes de negocio, revisión especializada de 22 tipos de contrato, recordatorios de renovación y borradores redline de revisión con control de cambios de Word. Los entregables formales de Feishu deben incluir la lista de tarjetas de problemas y el diagrama de flujo del negocio; cuando el usuario elige el modo de revisión, el contrato original se trata como fuente de solo lectura, se genera primero redline-plan.json y luego se ejecuta scripts/redline/apply_redline_plan.py para producir un documento Word de revisión con cambios registrados reales y comentarios, completando el QA del redline. Los demás registros se guardan como Markdown local." }, "fr": { "name": "pa1nrui1-contract-review", "summary": "Réalise une revue approfondie des contrats fondée sur la méthode de revue en quatre étapes de Zhang Haiyan plus la méthode d'analyse à trois perspectives de He Li/Chang Jinguang. Prend en charge la vérification d'entreprises, la détection automatique multilingue, les diagrammes de flux métier Mermaid, les avis de revue en deux étapes, les aperçus du contrat, les numéros de contrat lisibles par le client, les résumés métier, la revue spécialisée de 22 types de contrats, les rappels de renouvellement et les brouillons redline de revue en mode révisions de Word. Les livrables formels Feishu doivent inclure la liste des fiches de problèmes et le diagramme de flux métier ; lorsque l'utilisateur choisit le mode révision, le contrat original est traité comme source en lecture seule, redline-plan.json est d'abord généré, puis scripts/redline/apply_redline_plan.py est exécuté pour produire un document Word de revue avec modifications réelles suivies et commentaires, et achever le QA du redline. Les autres enregistrements sont sauvegardés en Markdown local." }, "ja": { "name": "pa1nrui1-contract-review", "summary": "張海燕の四ステップ審査法と何力/常金光の三観点分析法に基づき、契約の詳細審査を行います。企業照会、多言語自動検出、Mermaid業務フロー図、二段階審査意見、契約概要、顧客可読の契約番号、業務サマリー、22種類の契約特別審査、更新リマインダー、Wordの変更履歴モードによるレッドライン審査原稿に対応します。Feishuの正式な成果物には問題カードリストと業務フロー図を含めなければなりません。ユーザーが修正モードを選択した場合、元の契約を読み取り専用ソースとしてredline-plan.jsonを先に生成し、次にscripts/redline/apply_redline_plan.pyを実行して、実際の変更履歴とコメント付きのWord審査修正稿を生成し、レッドラインQAを完了します。その他の記録はローカルのMarkdownとして保存されます。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:contract-drafting", "slug": "skill-contract-drafting", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "The single entry point for contract drafting: generating contracts, preparing agreements, writing contracts, forming contract text from transaction arrangements, multi-round revisions, and archiving the final version. Based on the three-perspective four-step method and a user-configurable local cont" }, "human": { "zh-CN": { "name": "contract-drafting", "summary": "合同起草、生成合同、拟定协议、写合同、根据交易安排形成合同文本、多轮修订和最终版归档的唯一入口。基于三观四步法与用户可选配置的本地合同模板库，支持有模板时定制化起草、无模板时独立起草；输出合同草案，并同步生成客户可读合同编号、合同概览、业务摘要、版本记录、修订历史和续约提醒信息。" }, "en": { "name": "contract-drafting", "summary": "The single entry point for contract drafting: generating contracts, preparing agreements, writing contracts, forming contract text from transaction arrangements, multi-round revisions, and archiving the final version. Based on the three-perspective four-step method and a user-configurable local contract template library, it supports customized drafting with a template and standalone drafting without one; outputs a contract draft, and simultaneously generates a client-readable contract number, contract overview, business summary, version records, revision history, and renewal reminder information." }, "es": { "name": "contract-drafting", "summary": "El punto de entrada único para la redacción de contratos: generar contratos, preparar acuerdos, redactar contratos, formar el texto contractual a partir de acuerdos de transacción, revisiones multirronda y archivar la versión final. Basado en el método de tres perspectivas y cuatro pasos y en una biblioteca local de plantillas de contratos configurable por el usuario; admite redacción personalizada con plantilla y redacción autónoma sin ella. Entrega un borrador de contrato y genera simultáneamente un número de contrato legible para el cliente, un resumen del contrato, un resumen de negocio, registros de versión, historial de revisiones e información de recordatorio de renovación." }, "fr": { "name": "contract-drafting", "summary": "Le point d'entrée unique pour la rédaction de contrats : générer des contrats, préparer des accords, rédiger des contrats, former le texte contractuel à partir d'accords de transaction, révisions multi-tours et archivage de la version finale. Fondé sur la méthode des trois perspectives en quatre étapes et sur une bibliothèque locale de modèles de contrats configurable par l'utilisateur ; prend en charge la rédaction personnalisée avec modèle et la rédaction autonome sans modèle. Produit un projet de contrat et génère simultanément un numéro de contrat lisible par le client, un aperçu du contrat, un résumé métier, des enregistrements de version, un historique des révisions et des informations de rappel de renouvellement." }, "ja": { "name": "contract-drafting", "summary": "契約起草、契約生成、協定の作成、契約の執筆、取引条件に基づく契約テキストの作成、複数回の修正、最終版のアーカイブを担う唯一の入口です。三観点四ステップ法と、ユーザーが選択設定できるローカル契約テンプレートライブラリに基づき、テンプレートがある場合はカスタマイズ起草、ない場合は独立起草に対応します。契約ドラフトを出力し、同時に顧客可読の契約番号、契約概要、業務サマリー、バージョン記録、修正履歴、更新リマインダー情報を生成します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:engagement-contract-management", "slug": "skill-engagement-contract-management", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Assists lawyers with signing engagement contracts and archiving case files. The signing stage covers confirmation of the fee arrangement, confirmation of the scope of authorization, and generation of the full set of engagement documents; the archiving stage covers generation of litigation tracking r" }, "human": { "zh-CN": { "name": "engagement-contract-management", "summary": "协助律师完成委托合同签订和案件归档。签订阶段覆盖收费方式确认、授权范围确认、全套委托文书生成；归档阶段覆盖诉讼跟踪记录表生成和客户档案更新。当用户发送客户编号并提及\"签订合同\"、\"案件建档\"等指令时触发使用。" }, "en": { "name": "engagement-contract-management", "summary": "Assists lawyers with signing engagement contracts and archiving case files. The signing stage covers confirmation of the fee arrangement, confirmation of the scope of authorization, and generation of the full set of engagement documents; the archiving stage covers generation of litigation tracking record sheets and updating client files. Triggers when the user sends a client number and mentions instructions such as \"sign the contract\" or \"create the case file\"." }, "es": { "name": "engagement-contract-management", "summary": "Asiste a los abogados en la firma de contratos de encargo y en el archivo de expedientes. La etapa de firma cubre la confirmación del modo de honorarios, la confirmación del alcance de la autorización y la generación del juego completo de documentos de encargo; la etapa de archivo cubre la generación de la tabla de seguimiento del litigio y la actualización de los archivos del cliente. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como «firmar el contrato» o «crear el expediente del caso»." }, "fr": { "name": "engagement-contract-management", "summary": "Aide les avocats à conclure les contrats de mandat et à archiver les dossiers. La phase de signature couvre la confirmation du mode d'honoraires, la confirmation du périmètre d'autorisation et la génération du jeu complet de documents de mandat ; la phase d'archivage couvre la génération du tableau de suivi du contentieux et la mise à jour des dossiers clients. Se déclenche lorsque l'utilisateur envoie un numéro de client et mentionne des instructions telles que « signer le contrat » ou « créer le dossier de l'affaire »." }, "ja": { "name": "engagement-contract-management", "summary": "弁護士の委任契約の締結と案件のアーカイブを支援します。締結段階では報酬方式の確認、委任範囲の確認、委任書類一式の生成をカバーし、アーカイブ段階では訴訟追跡記録表の生成と顧客ファイルの更新をカバーします。ユーザーが顧客番号を送り、「契約の締結」「案件のファイル作成」などの指示に言及したときに起動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:prosecution-review-stage-defense", "slug": "skill-prosecution-review-stage-defense", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "A professional skill module covering all duties of the defense lawyer at the review-and-prosecution stage, including guidance on exercising seven core rights — case-file review, client meetings, investigation and evidence collection, evidence verification, expression of opinions, application for exc" }, "human": { "zh-CN": { "name": "prosecution-review-stage-defense", "summary": "覆盖审查起诉阶段辩护律师全部工作职责的专业技能模块，包括阅卷权、会见权、调查取证权、 证据核实权、意见表达权、非法证据排除申请权、控告申诉权等七类核心权利的行使指导， 以及法律文书生成、量刑预判、与检察机关沟通策略等全流程工作规范。" }, "en": { "name": "prosecution-review-stage-defense", "summary": "A professional skill module covering all duties of the defense lawyer at the review-and-prosecution stage, including guidance on exercising seven core rights — case-file review, client meetings, investigation and evidence collection, evidence verification, expression of opinions, application for exclusion of illegally obtained evidence, and complaint and petition rights — plus full-workflow standards such as legal document generation, sentencing prediction, and communication strategy with the procuratorate." }, "es": { "name": "prosecution-review-stage-defense", "summary": "Módulo de habilidad profesional que cubre todas las funciones del abogado defensor en la etapa de revisión y acusación, incluyendo la orientación para ejercer siete derechos centrales —revisión del expediente, entrevistas con el cliente, investigación y obtención de pruebas, verificación de pruebas, expresión de opiniones, solicitud de exclusión de pruebas obtenidas ilícitamente y derechos de denuncia y reclamación— además de normas de trabajo de flujo completo como la generación de documentos legales, la predicción de pena y la estrategia de comunicación con la fiscalía." }, "fr": { "name": "prosecution-review-stage-defense", "summary": "Module de compétence professionnel couvrant l'ensemble des missions de l'avocat de la défense au stade de l'examen avant poursuites, incluant des conseils sur l'exercice de sept droits fondamentaux — accès au dossier, entretiens avec le client, enquête et collecte de preuves, vérification des preuves, expression d'avis, demande d'exclusion de preuves obtenues illicitement et droits de plainte et de réclamation — ainsi que des normes de travail de flux complet telles que la génération de documents juridiques, la prédiction de la peine et la stratégie de communication avec le parquet." }, "ja": { "name": "prosecution-review-stage-defense", "summary": "審査起訴段階の弁護人の全職責をカバーする専門スキルモジュール。訴訟記録の閲覧、接見、調査・証拠収集、証拠の検証、意見表明、違法収集証拠排除の申立て、告訴・申立てという7種類の核心的権利の行使ガイダンスに加え、法律文書の生成、量刑予測、検察機関とのコミュニケーション戦略など、全フローの作業規範を含みます。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:trial-report-generator", "slug": "skill-trial-report-generator", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Extracts elements from existing case materials (complaints, defenses, hearing minutes, legal analyses, etc.) to generate a draft Word document of the court's internal trial report. Links to the master controller's current-matter records and retrospective ledger. Triggers when the user sends a client" }, "human": { "zh-CN": { "name": "trial-report-generator", "summary": "从已有案件材料（起诉状、答辩状、庭审纪要、法律分析等）中提取要素，生成法院内部审理报告Word文书初稿。按总控当前事项记录和复盘台账衔接。当用户发送客户编号并提及\"审理报告\"等指令时触发使用。" }, "en": { "name": "trial-report-generator", "summary": "Extracts elements from existing case materials (complaints, defenses, hearing minutes, legal analyses, etc.) to generate a draft Word document of the court's internal trial report. Links to the master controller's current-matter records and retrospective ledger. Triggers when the user sends a client number and mentions instructions such as \"trial report\"." }, "es": { "name": "trial-report-generator", "summary": "Extrae elementos de los materiales existentes del caso (demandas, contestaciones, actas de audiencia, análisis jurídicos, etc.) para generar un borrador de documento Word del informe interno de enjuiciamiento del tribunal. Se enlaza con los registros de asuntos actuales del controlador maestro y el libro de retrospectiva. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como «informe de enjuiciamiento»." }, "fr": { "name": "trial-report-generator", "summary": "Extrait des éléments des pièces existantes du dossier (actes de saisine, conclusions, procès-verbaux d'audience, analyses juridiques, etc.) pour générer un brouillon de document Word du rapport d'examen interne du tribunal. Se relie aux enregistrements des affaires en cours du contrôleur maître et au registre de rétrospection. Se déclenche lorsque l'utilisateur envoie un numéro de client et mentionne des instructions telles que « rapport d'examen »." }, "ja": { "name": "trial-report-generator", "summary": "既存の案件資料（起訴状、答弁状、公判記録、法律分析など）から要素を抽出し、裁判所内部の審理報告書のWord文書初稿を生成します。マスターコントロールの現在事項記録および振り返り台帳と連携します。ユーザーが顧客番号を送り、「審理報告書」などの指示に言及したときに起動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:advertising-compliance-review", "slug": "skill-advertising-compliance-review", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Advertising compliance review skill for checking whether advertising materials comply with China's Advertising Law and related regulations. Applicable scenarios: (1) the user submits ad copy or ad materials requesting compliance review; (2) the user mentions keywords such as \"ad review\", \"ad complia" }, "human": { "zh-CN": { "name": "advertising-compliance-review", "summary": "广告合规审核技能，用于审核广告素材是否符合中国广告法及相关法规。适用场景：(1) 用户提交广告文案、广告素材要求合规审核时；(2) 用户提到\"广告审核\"\"广告合规\"\"广告法审查\"等关键词时；(3) 用户要求检查广告内容是否存在违法违规风险时；(4) 用户提交房地产、食品、医疗、药品、互联网等行业广告要求专项审核时。审核依据涵盖《广告法》《反不正当竞争法》及行业专项法规。" }, "en": { "name": "advertising-compliance-review", "summary": "Advertising compliance review skill for checking whether advertising materials comply with China's Advertising Law and related regulations. Applicable scenarios: (1) the user submits ad copy or ad materials requesting compliance review; (2) the user mentions keywords such as \"ad review\", \"ad compliance\", or \"Advertising Law review\"; (3) the user asks to check whether ad content carries legal or regulatory violation risks; (4) the user submits ads from industries such as real estate, food, healthcare, pharmaceuticals, or the internet for specialized review. The review basis covers the Advertising Law, the Anti-Unfair Competition Law, and industry-specific regulations." }, "es": { "name": "advertising-compliance-review", "summary": "Habilidad de revisión de cumplimiento publicitario para comprobar si los materiales publicitarios cumplen la Ley de Publicidad de China y las normativas relacionadas. Escenarios aplicables: (1) el usuario envía textos o materiales publicitarios solicitando revisión de cumplimiento; (2) el usuario menciona palabras clave como «revisión publicitaria», «cumplimiento publicitario» o «revisión de la Ley de Publicidad»; (3) el usuario pide comprobar si el contenido publicitario presenta riesgos de infracción legal o reglamentaria; (4) el usuario envía anuncios de sectores como inmobiliario, alimentación, salud, farmacéutico o internet para revisión especializada. La base de revisión cubre la Ley de Publicidad, la Ley contra la Competencia Desleal y las normativas sectoriales específicas." }, "fr": { "name": "advertising-compliance-review", "summary": "Compétence de revue de conformité publicitaire pour vérifier si les supports publicitaires sont conformes à la loi chinoise sur la publicité et aux réglementations associées. Scénarios applicables : (1) l'utilisateur soumet des textes ou supports publicitaires pour revue de conformité ; (2) l'utilisateur mentionne des mots-clés tels que « revue publicitaire », « conformité publicitaire » ou « revue au titre de la loi sur la publicité » ; (3) l'utilisateur demande de vérifier si le contenu publicitaire présente des risques de violation légale ou réglementaire ; (4) l'utilisateur soumet des publicités de secteurs tels que l'immobilier, l'alimentaire, la santé, le pharmaceutique ou l'internet pour une revue spécialisée. La base de revue couvre la loi sur la publicité, la loi contre la concurrence déloyale et les réglementations sectorielles spécifiques." }, "ja": { "name": "advertising-compliance-review", "summary": "広告コンプライアンス審査スキル。広告素材が中国広告法および関連法規に適合しているかを審査します。適用シーン：(1) ユーザーが広告コピー・広告素材を提出してコンプライアンス審査を求める場合；(2) ユーザーが「広告審査」「広告コンプライアンス」「広告法審査」などのキーワードに言及する場合；(3) 広告コンテンツに法令違反リスクがないかの確認を求める場合；(4) 不動産、食品、医療、医薬品、インターネットなどの業界の広告について特別審査を求める場合。審査の根拠は『広告法』『反不正当競争法』および業界特有の法規をカバーします。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:pretrial-preparation", "slug": "skill-pretrial-preparation", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Pretrial preparation. Based on preceding documents such as complaints, legal analyses, litigation visualizations, case compilations, and cross-examination opinions, it automatically summarizes disputed issues and generates issue-oriented trial outlines, evidence-challenge attack-and-defense matrices" }, "human": { "zh-CN": { "name": "pretrial-preparation", "summary": "庭前准备。基于起诉状、法律分析、诉讼可视化、案例汇编、质证意见等前置文档，自动归纳争议焦点，生成争点导向型庭审提纲、质证攻防矩阵、举证策略规划表、开庭陈述提纲、法庭辩论提纲、代理词和诉讼法律服务方案。当用户发送客户编号并提及\"庭前准备\"、\"庭审提纲\"、\"代理词\"、\"诉讼方案\"等指令时触发使用。" }, "en": { "name": "pretrial-preparation", "summary": "Pretrial preparation. Based on preceding documents such as complaints, legal analyses, litigation visualizations, case compilations, and cross-examination opinions, it automatically summarizes disputed issues and generates issue-oriented trial outlines, evidence-challenge attack-and-defense matrices, evidence presentation strategy plans, opening statement outlines, courtroom debate outlines, representation statements, and litigation legal service plans. Triggers when the user sends a client number and mentions instructions such as \"pretrial preparation\", \"trial outline\", \"representation statement\", or \"litigation plan\"." }, "es": { "name": "pretrial-preparation", "summary": "Preparación del juicio. Con base en documentos previos como demandas, análisis jurídicos, visualizaciones del litigio, recopilaciones de casos y opiniones de impugnación de pruebas, resume automáticamente los puntos en disputa y genera esquemas de audiencia orientados a los puntos en disputa, matrices de ataque y defensa de impugnación de pruebas, planes de estrategia probatoria, esquemas de alegatos de apertura, esquemas de debate judicial, alegatos de representación y planes de servicios jurídicos de litigio. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como «preparación del juicio», «esquema de audiencia», «alegato» o «plan de litigio»." }, "fr": { "name": "pretrial-preparation", "summary": "Préparation de l'audience. À partir des documents en amont tels que actes de saisine, analyses juridiques, visualisations du contentieux, compilations de jurisprudence et avis de contestation des preuves, synthétise automatiquement les points de litige et génère des plans d'audience orientés vers les points de litige, des matrices d'attaque-défense de contestation des preuves, des tableaux de stratégie probatoire, des plans de plaidoirie d'ouverture, des plans de débat judiciaire, des conclusions de représentation et des propositions de services juridiques contentieux. Se déclenche lorsque l'utilisateur envoie un numéro de client et mentionne des instructions telles que « préparation de l'audience », « plan d'audience », « conclusions » ou « plan de contentieux »." }, "ja": { "name": "pretrial-preparation", "summary": "公判前準備。起訴状、法律分析、訴訟ビジュアライゼーション、判例集、証拠質疑意見などの事前資料に基づき、争点を自動的に整理し、争点指向型の公判アウトライン、質疑の攻防マトリクス、立証戦略計画表、冒頭陳述アウトライン、法廷弁論アウトライン、代理意見書、訴訟リーガルサービスプランを生成します。ユーザーが顧客番号を送り、「公判前準備」「公判アウトライン」「代理意見書」「訴訟プラン」などの指示に言及したときに起動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:trial-and-post-trial-work", "slug": "skill-trial-and-post-trial-work", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Systematically supports the full workflow: pre-hearing check reminders, hearing minutes records, post-hearing review analysis, drafting of post-hearing representation statements, supplementary material management, and communication with clients and judges. Triggers when the user sends a client numbe" }, "human": { "zh-CN": { "name": "trial-and-post-trial-work", "summary": "系统化支持庭审前检查提醒、庭审纪要记录、庭审复盘分析、庭后代理词撰写、补充材料管理、与当事人和法官沟通等全流程工作。当用户发送客户编号并提及\"开庭\"、\"庭审\"、\"庭后\"、\"代理词\"、\"复盘\"、\"庭审纪要\"、\"法官沟通\"等指令时触发使用。" }, "en": { "name": "trial-and-post-trial-work", "summary": "Systematically supports the full workflow: pre-hearing check reminders, hearing minutes records, post-hearing review analysis, drafting of post-hearing representation statements, supplementary material management, and communication with clients and judges. Triggers when the user sends a client number and mentions instructions such as \"court session\", \"hearing\", \"post-hearing\", \"representation statement\", \"review\", \"hearing minutes\", or \"judge communication\"." }, "es": { "name": "trial-and-post-trial-work", "summary": "Apoya sistemáticamente el flujo completo: recordatorios de verificación previa a la audiencia, registro de actas de audiencia, análisis de retrospectiva de la audiencia, redacción de alegatos posteriores, gestión de material complementario y comunicación con clientes y jueces. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como «audiencia», «juicio», «posjuicio», «alegato», «retrospectiva», «acta de audiencia» o «comunicación con el juez»." }, "fr": { "name": "trial-and-post-trial-work", "summary": "Soutient systématiquement le flux complet : rappels de vérification avant audience, enregistrement des procès-verbaux d'audience, analyse de rétrospection d'audience, rédaction des conclusions postérieures, gestion des pièces complémentaires et communication avec les clients et les juges. Se déclenche lorsque l'utilisateur envoie un numéro de client et mentionne des instructions telles que « audience », « procès », « après l'audience », « conclusions », « rétrospection », « procès-verbal d'audience » ou « communication avec le juge »." }, "ja": { "name": "trial-and-post-trial-work", "summary": "公判前のチェックリマインダー、公判記録の作成、公判の振り返り分析、公判後の代理意見書の執筆、補足資料の管理、当事者や裁判官とのコミュニケーションなど、全フローの業務を体系的に支援します。ユーザーが顧客番号を送り、「開廷」「公判」「公判後」「代理意見書」「振り返り」「公判記録」「裁判官との連絡」などの指示に言及したときに起動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:juvenile-criminal-cases", "slug": "skill-juvenile-criminal-cases", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Specialized defense skill for juvenile criminal cases, covering the education-influence-rescue principle for juvenile criminal cases, social investigation, non-arrest applications, conditional non-prosecution, criminal record sealing, non-public trials, presence of legal representatives, and informa" }, "human": { "zh-CN": { "name": "juvenile-criminal-cases", "summary": "未成年人刑事案件专项辩护Skill，覆盖未成年人刑事案件的教育感化挽救原则、社会调查、不逮捕申请、附条件不起诉、犯罪记录封存、不公开审理、法定代理人到场和信息保护等工作。当用户提及\"未成年人\"、\"少年犯\"、\"社会调查\"、\"附条件不起诉\"、\"犯罪记录封存\"、\"不公开审理\"等关键词时触发使用。" }, "en": { "name": "juvenile-criminal-cases", "summary": "Specialized defense skill for juvenile criminal cases, covering the education-influence-rescue principle for juvenile criminal cases, social investigation, non-arrest applications, conditional non-prosecution, criminal record sealing, non-public trials, presence of legal representatives, and information protection. Triggers when the user mentions keywords such as \"juvenile\", \"juvenile offender\", \"social investigation\", \"conditional non-prosecution\", \"criminal record sealing\", or \"non-public trial\"." }, "es": { "name": "juvenile-criminal-cases", "summary": "Habilidad de defensa especializada en causas penales de menores, que cubre el principio de educación-transformación-rescate en causas penales de menores, la investigación social, las solicitudes de no detención, la no acusación condicional, el sellado de antecedentes penales, los juicios no públicos, la presencia de los representantes legales y la protección de la información. Se activa cuando el usuario menciona palabras clave como «menor», «delincuente juvenil», «investigación social», «no acusación condicional», «sellado de antecedentes penales» o «juicio no público»." }, "fr": { "name": "juvenile-criminal-cases", "summary": "Compétence de défense spécialisée dans les affaires pénales de mineurs, couvrant le principe éducation-transformation-rescue dans les affaires pénales de mineurs, l'enquête sociale, les demandes de non-placement en détention, la non-poursuite conditionnelle, le scellement du casier judiciaire, les audiences à huis clos, la présence des représentants légaux et la protection des informations. Se déclenche lorsque l'utilisateur mentionne des mots-clés tels que « mineur », « délinquant juvénile », « enquête sociale », « non-poursuite conditionnelle », « scellement du casier judiciaire » ou « huis clos »." }, "ja": { "name": "juvenile-criminal-cases", "summary": "少年刑事事件に特化した弁護スキル。少年刑事事件における教育・感化・救済の原則、社会調査、逮捕しない処分の申立て、付条件不起訴、犯罪記録の封存、非公開審理、法定代理人の立会い、情報保護などの業務をカバーします。ユーザーが「未成年」「少年犯」「社会調査」「付条件不起訴」「犯罪記録の封存」「非公開審理」などのキーワードに言及したときに起動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:case-intake-and-engagement", "slug": "skill-case-intake-and-engagement", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Skill for taking on and engaging criminal defense cases. Assists lawyers with initial reception, case assessment, risk disclosure, engagement formalities, and case information collection for criminal cases; generates standardized engagement documents and establishes case files. Triggers when the use" }, "human": { "zh-CN": { "name": "case-intake-and-engagement", "summary": "刑事辩护案件承接与委托Skill。协助律师完成刑事案件的初次接待、案件评估、风险告知、委托手续办理及案件信息采集，生成标准化委托文书并建立案件档案。当用户发送客户编号并提及\"刑事案件承接\"、\"刑事委托\"、\"刑事接案\"等指令时触发使用。" }, "en": { "name": "case-intake-and-engagement", "summary": "Skill for taking on and engaging criminal defense cases. Assists lawyers with initial reception, case assessment, risk disclosure, engagement formalities, and case information collection for criminal cases; generates standardized engagement documents and establishes case files. Triggers when the user sends a client number and mentions instructions such as \"take on a criminal case\", \"criminal engagement\", or \"criminal case intake\"." }, "es": { "name": "case-intake-and-engagement", "summary": "Habilidad para la aceptación y contratación de casos de defensa penal. Asiste al abogado en la recepción inicial, la evaluación del caso, la advertencia de riesgos, las formalidades de encargo y la recopilación de información del caso en causas penales; genera documentos de encargo estandarizados y crea el expediente del caso. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como «aceptar un caso penal», «encargo penal» o «recepción de un caso penal»." }, "fr": { "name": "case-intake-and-engagement", "summary": "Compétence pour la prise en charge et le mandatement d'affaires de défense pénale. Aide l'avocat dans l'accueil initial, l'évaluation de l'affaire, l'information sur les risques, les formalités de mandat et la collecte d'informations de l'affaire pénale ; génère des documents de mandat standardisés et établit le dossier de l'affaire. Se déclenche lorsque l'utilisateur envoie un numéro de client et mentionne des instructions telles que « prise en charge d'une affaire pénale », « mandat pénal » ou « acceptation d'une affaire pénale »." }, "ja": { "name": "case-intake-and-engagement", "summary": "刑事弁護案件の受任と委任のためのスキル。刑事事件の初回面接、案件評価、リスク告知、委任手続きの処理、案件情報の収集を弁護士に支援し、標準化された委任書類を生成して案件ファイルを作成します。ユーザーが顧客番号を送り、「刑事事件の受任」「刑事委任」「刑事の受案」などの指示に言及したときに起動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:client-communication-records", "slug": "skill-client-communication-records", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Creates communication record documents for civil and commercial litigation cases. Following the requirements of the Litigation Refinement manual, it records the content of the first in-depth client communication in a dialogue-transcript format and finally generates the case communication record as a" }, "human": { "zh-CN": { "name": "client-communication-records", "summary": "制作民商事诉讼案件沟通记录文档，依据《诉讼精细化》手册要求，采用对话式笔录格式记录与客户首次深度沟通的内容，最终生成Word格式的案件沟通记录。当用户发送客户编号并提及\"案件沟通记录\"、\"制作沟通记录\"或类似指令时触发使用。" }, "en": { "name": "client-communication-records", "summary": "Creates communication record documents for civil and commercial litigation cases. Following the requirements of the Litigation Refinement manual, it records the content of the first in-depth client communication in a dialogue-transcript format and finally generates the case communication record as a Word document. Triggers when the user sends a client number and mentions instructions such as \"case communication record\" or \"create a communication record\"." }, "es": { "name": "client-communication-records", "summary": "Crea documentos de registro de comunicaciones para casos de litigio civil y mercantil. Siguiendo los requisitos del manual Refinamiento del Litigio, registra el contenido de la primera comunicación en profundidad con el cliente en formato de transcripción dialogada y finalmente genera el registro de comunicaciones del caso como documento Word. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como «registro de comunicaciones del caso» o «elaborar el registro de comunicaciones»." }, "fr": { "name": "client-communication-records", "summary": "Crée des documents de registre de communication pour les dossiers de contentieux civil et commercial. Conformément aux exigences du manuel Raffinement du contentieux, il enregistre le contenu de la première communication approfondie avec le client sous forme de transcription dialoguée et génère finalement le registre de communication de l'affaire au format Word. Se déclenche lorsque l'utilisateur envoie un numéro de client et mentionne des instructions telles que « registre de communication de l'affaire » ou « établir un registre de communication »." }, "ja": { "name": "client-communication-records", "summary": "民事・商事訴訟案件のコミュニケーション記録文書を作成します。『訴訟精緻化』マニュアルの要件に従い、顧客との初回の詳細なコミュニケーション内容を対話型の筆録形式で記録し、最終的にWord形式の案件コミュニケーション記録を生成します。ユーザーが顧客番号を送り、「案件コミュニケーション記録」「コミュニケーション記録の作成」などの指示に言及したときに起動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:case-discussion-outline", "slug": "skill-case-discussion-outline", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "An AI-automated deep case analysis tool that integrates preceding materials to perform four-dimension self-question-and-answer analysis and a three-party moot court rehearsal, ultimately generating a case outline. Triggers when the user mentions instructions such as \"case discussion\", \"case outline\"" }, "human": { "zh-CN": { "name": "case-discussion-outline", "summary": "AI自动完成的深度案件分析工具，整合前序材料进行四维度自问自答分析、三方模拟法庭演练，最终生成案件提纲。当用户提及\"案件讨论\"、\"案件提纲\"、\"案件分析\"、\"模拟法庭\"、\"案件研讨\"等指令时触发使用。立案完成后主动提醒律师执行。" }, "en": { "name": "case-discussion-outline", "summary": "An AI-automated deep case analysis tool that integrates preceding materials to perform four-dimension self-question-and-answer analysis and a three-party moot court rehearsal, ultimately generating a case outline. Triggers when the user mentions instructions such as \"case discussion\", \"case outline\", \"case analysis\", \"moot court\", or \"case review\". Proactively reminds the lawyer to run it after case filing is complete." }, "es": { "name": "case-discussion-outline", "summary": "Herramienta de análisis profundo de casos automatizada por IA que integra los materiales previos para realizar un análisis de auto-preguntas y respuestas en cuatro dimensiones y una simulación de juicio con tres partes, generando finalmente un esquema del caso. Se activa cuando el usuario menciona instrucciones como «discusión del caso», «esquema del caso», «análisis del caso», «juicio simulado» o «estudio del caso». Recuerda proactivamente al abogado ejecutarlo tras completar la presentación de la demanda." }, "fr": { "name": "case-discussion-outline", "summary": "Outil d'analyse approfondie d'affaires automatisé par IA qui intègre les pièces en amont pour réaliser une analyse auto-questionnement-réponse à quatre dimensions et une plaidoirie simulée à trois parties, générant finalement le plan de l'affaire. Se déclenche lorsque l'utilisateur mentionne des instructions telles que « discussion de l'affaire », « plan de l'affaire », « analyse de l'affaire », « procès simulé » ou « étude de l'affaire ». Rappelle proactivement à l'avocat de l'exécuter une fois l'enregistrement de l'affaire achevé." }, "ja": { "name": "case-discussion-outline", "summary": "AIが自動で行う案件の詳細分析ツール。先行資料を統合し、4次元の自問自答分析と3者模擬法廷のリハーサルを行い、最終的に案件アウトラインを生成します。ユーザーが「案件ディスカッション」「案件アウトライン」「案件分析」「模擬法廷」「案件検討」などの指示に言及したときに起動します。立案完了後に弁護士へ実行を能動的にリマインドします。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:death-penalty-defense", "slug": "skill-death-penalty-defense", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Specialized criminal defense skill for death penalty and death-with-reprieve cases, covering identification of cases that may carry the death penalty, first-instance life-preservation defense, mandatory second-instance court hearings, victim reconciliation, death penalty review defense, evidence sta" }, "human": { "zh-CN": { "name": "death-penalty-defense", "summary": "死刑案件及死缓案件刑事辩护专项Skill，覆盖可能判处死刑案件识别、一审保命辩护、二审强制开庭、被害人和解、死刑复核辩护、死刑案件证据标准和被害人代理注意事项。当用户提及\"死刑\"、\"死缓\"、\"死刑立即执行\"、\"死刑复核\"、\"核准死刑\"等关键词时触发使用。" }, "en": { "name": "death-penalty-defense", "summary": "Specialized criminal defense skill for death penalty and death-with-reprieve cases, covering identification of cases that may carry the death penalty, first-instance life-preservation defense, mandatory second-instance court hearings, victim reconciliation, death penalty review defense, evidence standards for death penalty cases, and cautions for victim representation. Triggers when the user mentions keywords such as \"death penalty\", \"death with reprieve\", \"immediate execution\", \"death penalty review\", or \"approval of death penalty\"." }, "es": { "name": "death-penalty-defense", "summary": "Habilidad de defensa penal especializada en casos de pena de muerte y de pena de muerte con suspensión, que cubre la identificación de casos que pueden acarrear pena de muerte, la defensa de preservación de la vida en primera instancia, la celebración obligatoria de vista en segunda instancia, la reconciliación con la víctima, la defensa en la revisión de la pena de muerte, los estándares probatorios en casos de pena de muerte y las precauciones para la representación de la víctima. Se activa cuando el usuario menciona palabras clave como «pena de muerte», «pena de muerte con suspensión», «ejecución inmediata», «revisión de la pena de muerte» o «aprobación de la pena de muerte»." }, "fr": { "name": "death-penalty-defense", "summary": "Compétence de défense pénale spécialisée dans les affaires de peine de mort et de peine de mort avec sursis, couvrant l'identification des affaires pouvant entraîner la peine de mort, la défense de préservation de la vie en première instance, l'audience obligatoire en appel, la réconciliation avec la victime, la défense en phase de révision de la peine de mort, les normes probatoires des affaires de peine de mort et les points de vigilance pour la représentation de la victime. Se déclenche lorsque l'utilisateur mentionne des mots-clés tels que « peine de mort », « peine de mort avec sursis », « exécution immédiate », « révision de la peine de mort » ou « homologation de la peine de mort »." }, "ja": { "name": "death-penalty-defense", "summary": "死刑事件および死刑緩期執行事件に特化した刑事弁護スキル。死刑に相当する可能性のある事件の識別、一審での生命維持弁護、二審の強制開廷、被害者との和解、死刑复核弁護、死刑事件の証拠基準、被害者代理の注意点をカバーします。ユーザーが「死刑」「死刑緩期執行」「死刑の直ちに執行」「死刑复核」「死刑の核准」などのキーワードに言及したときに起動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:civil-first-instance-litigation", "slug": "skill-civil-first-instance-litigation", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Professional workflow-orchestration skill for civil first-instance litigation and bankruptcy cases. Integrates 25 sub-skills: 14 civil litigation skills (legal consultation assistant, case communication records, preliminary legal analysis, engagement contract management, statute and case retrieval, " }, "human": { "zh-CN": { "name": "civil-first-instance-litigation", "summary": "民事一审诉讼与破产案件专业流程编排 Skill。整合25个子Skill：民事诉讼14个（法律咨询助手、案件沟通记录、初步法律分析、委托合同管理、法规案例检索、调查取证与证据管理、立案管理、诉讼可视化、案件讨论与提纲、诉讼文书起草、庭前准备、庭审与庭后工作、调解与和解、诉讼分析工具、结案归档）+ 破产案件10个（破产法律分析、破产申请与受理、管理人工作、财产调查与管理、债权申报与审查、债权人会议、重整程序、破产和解程序、破产清算、破产文书生成）。用于判断民事/破产案件阶段（S0-S16/SM/BP系列）、推荐下一步子Skill、提示关键节点确认和期限；事项建档、路径、文件读取、来源披露、OCR 校正和缺口归档由「法律工作总控」统一处理。" }, "en": { "name": "civil-first-instance-litigation", "summary": "Professional workflow-orchestration skill for civil first-instance litigation and bankruptcy cases. Integrates 25 sub-skills: 14 civil litigation skills (legal consultation assistant, case communication records, preliminary legal analysis, engagement contract management, statute and case retrieval, investigation and evidence collection and evidence management, case filing management, litigation visualization, case discussion and outline, litigation document drafting, pretrial preparation, trial and post-trial work, mediation and settlement, litigation analysis tools, and case closing and archiving) plus 10 bankruptcy skills (bankruptcy legal analysis, bankruptcy application and acceptance, administrator work, asset investigation and management, creditor claim registration and review, creditors' meetings, reorganization procedures, bankruptcy settlement procedures, bankruptcy liquidation, and bankruptcy document generation). Used to determine the civil/bankruptcy case stage (S0-S16/SM/BP series), recommend the next sub-skill, and flag key-node confirmations and deadlines; matter filing, paths, file reading, source disclosure, OCR correction, and gap archiving are handled uniformly by the \"legal work master controller\"." }, "es": { "name": "civil-first-instance-litigation", "summary": "Habilidad de orquestación de flujo profesional para el litigio civil de primera instancia y los casos concursales. Integra 25 subhabilidades: 14 de litigio civil (asistente de consultas jurídicas, registros de comunicación del caso, análisis jurídico preliminar, gestión de contratos de encargo, búsqueda de normas y jurisprudencia, investigación y obtención de pruebas y gestión de pruebas, gestión de presentación de la demanda, visualización del litigio, discusión y esquema del caso, redacción de documentos procesales, preparación del juicio, audiencia y trabajo posterior, mediación y transacción, herramientas de análisis del litigio y cierre y archivo del caso) más 10 de concursos (análisis jurídico concursal, solicitud y admisión del concurso, trabajo del administrador concursal, investigación y gestión de los bienes, declaración y examen de créditos, juntas de acreedores, procedimientos de reorganización, procedimientos de concordato concursal, liquidación concursal y generación de documentos concursales). Se usa para determinar la etapa del caso civil/concursal (serie S0-S16/SM/BP), recomendar la siguiente subhabilidad y señalar confirmaciones de nodos clave y plazos; el expediente del asunto, las rutas, la lectura de archivos, la divulgación de fuentes, la corrección OCR y el archivo de brechas se gestionan de forma unificada por el «controlador maestro de trabajo legal»." }, "fr": { "name": "civil-first-instance-litigation", "summary": "Compétence d'orchestration professionnelle de flux pour le contentieux civil de première instance et les dossiers de faillite. Intègre 25 sous-compétences : 14 de contentieux civil (assistant de consultation juridique, registres de communication de l'affaire, analyse juridique préliminaire, gestion des contrats de mandat, recherche de textes et de jurisprudence, enquête et collecte de preuves et gestion des preuves, gestion de l'enregistrement de l'affaire, visualisation du contentieux, discussion et plan de l'affaire, rédaction des actes de procédure, préparation de l'audience, audience et travail postérieur, médiation et transaction, outils d'analyse du contentieux et clôture et archivage de l'affaire) plus 10 de faillite (analyse juridique de la faillite, demande et ouverture de la procédure, travail de l'administrateur, enquête et gestion de l'actif, déclaration et examen des créances, assemblées des créanciers, procédures de redressement, procédures de concordat de faillite, liquidation de la faillite et génération des documents de faillite). Sert à déterminer la phase de l'affaire civile/de faillite (série S0-S16/SM/BP), recommander la sous-compétence suivante et signaler les confirmations de jalons clés et les échéances ; l'ouverture du dossier, les chemins, la lecture de fichiers, la divulgation des sources, la correction OCR et l'archivage des lacunes sont gérés uniformément par le « contrôleur maître du travail juridique »." }, "ja": { "name": "civil-first-instance-litigation", "summary": "民事一審訴訟と破産案件のためのプロフェッショナルなワークフローオーケストレーションスキル。25のサブスキルを統合します：民事訴訟14個（法律相談アシスタント、案件コミュニケーション記録、初期法律分析、委任契約管理、法令・判例検索、調査・証拠収集と証拠管理、提訴・受理管理、訴訟ビジュアライゼーション、案件ディスカッションとアウトライン、訴訟文書起草、公判前準備、公判と公判後業務、調停と和解、訴訟分析ツール、結案・アーカイブ）＋破産案件10個（破産法律分析、破産申立てと受理、管財人業務、財産調査と管理、債権届出と審査、債権者集会、更生手続き、破産和解手続き、破産清算、破産文書生成）。民事/破産案件の段階判定（S0-S16/SM/BPシリーズ）、次のサブスキルの推奨、重要ノードの確認と期限の提示に使用します。案件のファイル作成、パス、ファイル読み取り、ソースの開示、OCR補正、ギャップのアーカイブは「法律業務マスターコントロール」が一元的に処理します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:civil-judgment-drafting", "slug": "skill-civil-judgment-drafting", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "A first-instance ordinary-procedure civil judgment drafting tool. Extracts elements from existing case materials (complaints, defenses, hearing minutes, legal analyses, etc.) to generate a draft Word judgment that conforms to court document style standards. Links to the master controller's current-m" }, "human": { "zh-CN": { "name": "civil-judgment-drafting", "summary": "一审普通程序民事判决书生成工具。从已有案件材料（起诉状、答辩状、庭审纪要、法律分析等）提取要素，生成符合法院文书样式标准的Word判决书初稿。按总控当前事项记录和复盘台账衔接。当用户发送客户编号并提及\"判决书\"、\"民事判决书\"、\"起草判决书\"、\"模拟判决书\"等指令时触发使用。" }, "en": { "name": "civil-judgment-drafting", "summary": "A first-instance ordinary-procedure civil judgment drafting tool. Extracts elements from existing case materials (complaints, defenses, hearing minutes, legal analyses, etc.) to generate a draft Word judgment that conforms to court document style standards. Links to the master controller's current-matter records and retrospective ledger. Triggers when the user sends a client number and mentions instructions such as \"judgment\", \"civil judgment\", \"draft a judgment\", or \"simulated judgment\"." }, "es": { "name": "civil-judgment-drafting", "summary": "Herramienta de generación de sentencias civiles de primera instancia en procedimiento ordinario. Extrae elementos de los materiales existentes del caso (demandas, contestaciones, actas de audiencia, análisis jurídicos, etc.) para generar un borrador de sentencia en Word conforme a los estándares de estilo de los documentos judiciales. Se enlaza con los registros de asuntos actuales del controlador maestro y el libro de retrospectiva. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como «sentencia», «sentencia civil», «redactar la sentencia» o «sentencia simulada»." }, "fr": { "name": "civil-judgment-drafting", "summary": "Outil de rédaction de jugements civils de première instance en procédure ordinaire. Extrait des éléments des pièces existantes du dossier (actes de saisine, conclusions, procès-verbaux d'audience, analyses juridiques, etc.) pour générer un brouillon de jugement Word conforme aux normes de style des documents judiciaires. Se relie aux enregistrements des affaires en cours du contrôleur maître et au registre de rétrospection. Se déclenche lorsque l'utilisateur envoie un numéro de client et mentionne des instructions telles que « jugement », « jugement civil », « rédiger le jugement » ou « jugement simulé »." }, "ja": { "name": "civil-judgment-drafting", "summary": "一審通常手続きの民事判決書生成ツール。既存の案件資料（起訴状、答弁状、公判記録、法律分析など）から要素を抽出し、裁判所文書様式の基準に適合したWord判決書の初稿を生成します。マスターコントロールの現在事項記録および振り返り台帳と連携します。ユーザーが顧客番号を送り、「判決書」「民事判決書」「判決書の起草」「模擬判決書」などの指示に言及したときに起動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:legal-wiki-lookup", "slug": "skill-legal-wiki-lookup", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Searches Wiki knowledge pages by crime/charge, practice area, or keyword, returning knowledge cards or full-text content. Supports queries and cross-reference jumps across criminal charge knowledge, civil area knowledge, and case knowledge pages. Triggers when the user mentions keywords such as \"che" }, "human": { "zh-CN": { "name": "legal-wiki-lookup", "summary": "按罪名/领域/关键词检索 Wiki 知识页，返回知识卡片或全文内容。 支持刑事罪名知识、民事领域知识、案例知识页的查询与交叉引用跳转。 当用户提及\"查Wiki\"、\"知识库查询\"、\"查罪名\"、\"查法律知识\"、\"wiki查询\"等关键词时触发。" }, "en": { "name": "legal-wiki-lookup", "summary": "Searches Wiki knowledge pages by crime/charge, practice area, or keyword, returning knowledge cards or full-text content. Supports queries and cross-reference jumps across criminal charge knowledge, civil area knowledge, and case knowledge pages. Triggers when the user mentions keywords such as \"check the Wiki\", \"knowledge base query\", \"look up a charge\", \"look up legal knowledge\", or \"wiki query\"." }, "es": { "name": "legal-wiki-lookup", "summary": "Busca páginas de conocimiento del Wiki por delito, ámbito de práctica o palabra clave, y devuelve tarjetas de conocimiento o contenido completo. Admite consultas y saltos de referencia cruzada entre el conocimiento de delitos penales, el conocimiento del ámbito civil y las páginas de conocimiento de casos. Se activa cuando el usuario menciona palabras clave como «consultar el Wiki», «consulta de la base de conocimiento», «buscar un delito», «buscar conocimientos jurídicos» o «consulta wiki»." }, "fr": { "name": "legal-wiki-lookup", "summary": "Recherche les pages de connaissances du Wiki par infraction, domaine de pratique ou mot-clé, et renvoie des fiches de connaissances ou le contenu intégral. Prend en charge les requêtes et les sauts de références croisées entre les connaissances des infractions pénales, les connaissances du domaine civil et les pages de connaissances d'affaires. Se déclenche lorsque l'utilisateur mentionne des mots-clés tels que « consulter le Wiki », « requête dans la base de connaissances », « rechercher une infraction », « rechercher des connaissances juridiques » ou « requête wiki »." }, "ja": { "name": "legal-wiki-lookup", "summary": "罪名・分野・キーワードでWikiナレッジページを検索し、ナレッジカードまたは全文を返します。刑事罪名のナレッジ、民事分野のナレッジ、事例ナレッジページの照会と相互参照ジャンプに対応します。ユーザーが「Wikiを調べる」「ナレッジベース照会」「罪名を調べる」「法律知識を調べる」「Wiki照会」などのキーワードに言及したときに起動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:legal-consultation-assistant", "slug": "skill-legal-consultation-assistant", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Assists lawyers in handling legal consultations via WeChat and offline consultations, providing colloquial reply suggestions, intelligent follow-up question guidance, and soft-sell placement, and updating the retrospective ledger when the user explicitly requests it or the master controller confirms" }, "human": { "zh-CN": { "name": "legal-consultation-assistant", "summary": "协助律师处理微信渠道法律咨询和线下咨询，提供口语化回复建议、智能反问引导、软性销售植入，并在用户明确要求或总控确认需要时更新复盘台账。支持客户编号识别、咨询入口路由、线下咨询事实补全等功能。" }, "en": { "name": "legal-consultation-assistant", "summary": "Assists lawyers in handling legal consultations via WeChat and offline consultations, providing colloquial reply suggestions, intelligent follow-up question guidance, and soft-sell placement, and updating the retrospective ledger when the user explicitly requests it or the master controller confirms the need. Supports client number recognition, consultation entry routing, and offline consultation fact completion." }, "es": { "name": "legal-consultation-assistant", "summary": "Asiste a los abogados en la gestión de consultas jurídicas por WeChat y consultas presenciales, aportando sugerencias de respuesta coloquiales, guía inteligente de contra-preguntas e inserción suave de venta, y actualizando el libro de retrospectiva cuando el usuario lo solicita explícitamente o el controlador maestro confirma la necesidad. Admite el reconocimiento del número de cliente, el enrutamiento del canal de consulta y la completación de hechos en consultas presenciales." }, "fr": { "name": "legal-consultation-assistant", "summary": "Aide les avocats à traiter les consultations juridiques via WeChat et les consultations en présentiel, en fournissant des suggestions de réponses familières, un guidage intelligent par contre-questions et une intégration commerciale discrète, et en mettant à jour le registre de rétrospection lorsque l'utilisateur le demande explicitement ou que le contrôleur maître confirme la nécessité. Prend en charge la reconnaissance du numéro de client, le routage du canal de consultation et la complétion des faits des consultations en présentiel." }, "ja": { "name": "legal-consultation-assistant", "summary": "弁護士のWeChatチャネルおよび対面の法律相談を支援し、口語的な返信案の提案、インテリジェントな確認質問のガイド、ソフトセールスの組み込みを提供し、ユーザーが明示的に求めた場合またはマスターコントロールが必要と確認した場合に振り返り台帳を更新します。顧客番号の識別、相談入口のルーティング、対面相談の事実補完などの機能に対応します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:legal-workflow-controller", "slug": "skill-legal-workflow-controller", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "General entry-point skill for the legal folder. Provides semantic routing, case isolation, source disclosure, file-read re-checking, statute/Wiki verification, OCR correction, gap flagging, and contract preference learning for tasks such as legal consultation, case handling, contracts, product legal" }, "human": { "zh-CN": { "name": "legal-workflow-controller", "summary": "legal 文件夹通用入口 Skill。用于法律咨询、案件办理、合同、产品法务、监管合规、诉讼、刑辩、劳动争议、破产、合规、文书、检索等任务的语义路由、案件隔离、来源披露、文件读取复查、法规/Wiki 校验、OCR 校正、缺口提示和合同偏好学习。用户提出任何法律工作请求、客户编号、案件材料处理、法律文书生成或需要自动匹配 legal 子 Skill 时触发。" }, "en": { "name": "legal-workflow-controller", "summary": "General entry-point skill for the legal folder. Provides semantic routing, case isolation, source disclosure, file-read re-checking, statute/Wiki verification, OCR correction, gap flagging, and contract preference learning for tasks such as legal consultation, case handling, contracts, product legal affairs, regulatory compliance, litigation, criminal defense, labor disputes, bankruptcy, compliance, document drafting, and retrieval. Triggered when the user makes any legal work request, provides a client number, requests case material processing or legal document generation, or needs automatic matching to a legal sub-skill." }, "es": { "name": "legal-workflow-controller", "summary": "Skill de punto de entrada general para la carpeta legal. Proporciona enrutamiento semántico, aislamiento de casos, divulgación de fuentes, reverificación de lecturas de archivos, validación de leyes/Wiki, corrección de OCR, señalización de brechas y aprendizaje de preferencias contractuales para tareas como consultas legales, gestión de casos, contratos, asuntos legales de producto, cumplimiento regulatorio, litigios, defensa penal, conflictos laborales, quiebra, cumplimiento, redacción de documentos y búsquedas. Se activa cuando el usuario formula cualquier solicitud de trabajo legal, proporciona un número de cliente, pide procesar materiales del caso o generar documentos legales, o necesita el emparejamiento automático con un sub-skill legal." }, "fr": { "name": "legal-workflow-controller", "summary": "Skill de point d'entrée général du dossier legal. Assure le routage sémantique, l'isolation des dossiers, la divulgation des sources, la re-vérification des lectures de fichiers, la validation des lois/Wiki, la correction OCR, le signalement des lacunes et l'apprentissage des préférences contractuelles pour des tâches telles que consultation juridique, traitement de dossiers, contrats, affaires juridiques produit, conformité réglementaire, contentieux, défense pénale, litiges du travail, faillite, conformité, rédaction d'actes et recherche documentaire. Déclenché dès que l'utilisateur formule une demande de travail juridique, fournit un numéro de client, demande le traitement de pièces du dossier ou la génération de documents juridiques, ou a besoin du rattachement automatique à un sous-skill legal." }, "ja": { "name": "legal-workflow-controller", "summary": "legal フォルダの汎用エントリポイント skill。法律相談、案件処理、契約、製品法務、規制コンプライアンス、訴訟、刑事弁護、労務紛争、破産、コンプライアンス、文書作成、検索などのタスクに対して、セマンティックルーティング、案件隔離、出典開示、ファイル読み込み再確認、法令/Wiki 検証、OCR 補正、欠落の提示、契約プリファレンスの学習を提供します。ユーザーが法律業務のリクエスト、顧客番号、案件資料の処理、法律文書の生成を依頼した場合や、legal サブ skill の自動マッチングが必要な場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:legal-doc-pre-release-review", "slug": "skill-legal-doc-pre-release-review", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Triggered after a legal business skill generates the body text or element-based fields and before legal document templates are exported to a local Word file. Used to review draft.html/preflight-meta.json or complaint-data.json/fill-plan.json, the file-read re-check summary, the statute verification " }, "human": { "zh-CN": { "name": "legal-doc-pre-release-review", "summary": "在法律业务 Skill 生成正文或要素式字段后、法律文书模板与导出生成本地 Word 前触发。用于审查 draft.html/preflight-meta.json 或 complaint-data.json/fill-plan.json、读取复查摘要、法规校验摘要、来源边界和用户确认记录，决定是否可以进入正式 DOCX 导出；发现问题后必须闭环推进到业务 Skill 整改、用户确认或材料读取流程。" }, "en": { "name": "legal-doc-pre-release-review", "summary": "Triggered after a legal business skill generates the body text or element-based fields and before legal document templates are exported to a local Word file. Used to review draft.html/preflight-meta.json or complaint-data.json/fill-plan.json, the file-read re-check summary, the statute verification summary, source boundaries, and user confirmation records, and to decide whether formal DOCX export may proceed; when issues are found, closure must be driven through business-skill remediation, user confirmation, or the material re-reading flow." }, "es": { "name": "legal-doc-pre-release-review", "summary": "Se activa después de que un skill de negocio legal genera el cuerpo del texto o los campos por elementos y antes de que las plantillas de documentos legales se exporten a un archivo Word local. Sirve para revisar draft.html/preflight-meta.json o complaint-data.json/fill-plan.json, el resumen de reverificación de lecturas, el resumen de validación de leyes, los límites de las fuentes y los registros de confirmación del usuario, y decidir si se puede proceder a la exportación DOCX formal; si se detectan problemas, el ciclo debe cerrarse mediante la corrección por el skill de negocio, la confirmación del usuario o el flujo de relectura de materiales." }, "fr": { "name": "legal-doc-pre-release-review", "summary": "Déclenché après qu'un skill métier juridique a généré le corps du texte ou les champs par éléments, et avant l'export des modèles de documents juridiques vers un fichier Word local. Sert à examiner draft.html/preflight-meta.json ou complaint-data.json/fill-plan.json, le résumé de re-vérification des lectures, le résumé de validation des lois, les limites des sources et les enregistrements de confirmation de l'utilisateur, et à décider si l'export DOCX formel peut procéder ; en cas de problème, la boucle doit être fermée via la correction par le skill métier, la confirmation de l'utilisateur ou le flux de relecture des pièces." }, "ja": { "name": "legal-doc-pre-release-review", "summary": "法律業務 skill が本文または要素型フィールドを生成した後、法律文書テンプレートをローカルの Word ファイルに書き出す前に発動します。draft.html/preflight-meta.json または complaint-data.json/fill-plan.json、読み込み再確認サマリー、法令検証サマリー、出典境界、ユーザー確認記録を審査し、正式な DOCX 書き出しに進めてよいかを判定します。問題が見つかった場合は、業務 skill による修正、ユーザー確認、または資料再読み込みフローへ閉ループで引き継ぐ必要があります。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:legal-doc-templates-export", "slug": "skill-legal-doc-templates-export", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Uniformly handles documents, reports, checklists, transcripts, opinions, letters, contracts, and formal deliverables in legal work that ultimately need to be output as local Word (.docx) files. Invoked by mandatory routing from the legal work controller; business skills are responsible for the body " }, "human": { "zh-CN": { "name": "legal-doc-templates-export", "summary": "统一处理法律工作中最终需要输出本地 Word（.docx）的文书、报告、清单、笔录、意见书、函件、合同和正式交付文件。由法律工作总控强制路由调用；业务 Skill 负责正文和法律判断，本 Skill 负责选择格式 profile、接收语义 HTML 或要素式填充数据、导出 DOCX、结构体检和兜底模板。" }, "en": { "name": "legal-doc-templates-export", "summary": "Uniformly handles documents, reports, checklists, transcripts, opinions, letters, contracts, and formal deliverables in legal work that ultimately need to be output as local Word (.docx) files. Invoked by mandatory routing from the legal work controller; business skills are responsible for the body text and legal judgments, while this skill selects the format profile, receives semantic HTML or element-based fill data, exports the DOCX, performs structural checks, and applies fallback templates." }, "es": { "name": "legal-doc-templates-export", "summary": "Gestiona de manera uniforme los documentos, informes, listas de verificación, actas, dictámenes, cartas, contratos y entregables formales del trabajo legal que deban emitirse finalmente como archivos Word (.docx) locales. Se invoca mediante enrutamiento obligatorio desde el controlador del trabajo legal; los skills de negocio se encargan del cuerpo del texto y de los juicios jurídicos, mientras que este skill selecciona el perfil de formato, recibe HTML semántico o datos de relleno por elementos, exporta el DOCX, realiza verificaciones estructurales y aplica plantillas de respaldo." }, "fr": { "name": "legal-doc-templates-export", "summary": "Traite de manière uniforme les actes, rapports, listes de contrôle, procès-verbaux, avis, courriers, contrats et livrables formels du travail juridique qui doivent in fine être produits en fichiers Word (.docx) locaux. Invoqué par routage obligatoire depuis le contrôleur du travail juridique ; les skills métier prennent en charge le corps du texte et les appréciations juridiques, tandis que ce skill choisit le profil de format, reçoit du HTML sémantique ou des données de remplissage par éléments, exporte le DOCX, effectue les contrôles structurels et applique les modèles de repli." }, "ja": { "name": "legal-doc-templates-export", "summary": "法律業務において最終的にローカルの Word（.docx）として出力が必要な文書、報告書、チェックリスト、調書、意見書、書簡、契約、正式納品物を統一的に処理します。法律業務コントローラによる強制ルーティングで呼び出されます。本文と法的判断は業務 skill が担当し、本 skill はフォーマットプロファイルの選択、セマンティック HTML または要素型入力データの受け取り、DOCX の書き出し、構造チェック、フォールバックテンプレートを担います。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:legal-writing-deai-polish", "slug": "skill-legal-writing-deai-polish", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Detects and removes AI-style phrasing patterns in articles; used for writing polish, text optimization, and removing the AI tone." }, "human": { "zh-CN": { "name": "legal-writing-deai-polish", "summary": "检测并去除文章中的AI化表述模式，用于写作润色、文本优化、去AI腔。" }, "en": { "name": "legal-writing-deai-polish", "summary": "Detects and removes AI-style phrasing patterns in articles; used for writing polish, text optimization, and removing the AI tone." }, "es": { "name": "legal-writing-deai-polish", "summary": "Detecta y elimina patrones de redacción de estilo IA en los artículos; se usa para pulir escritos, optimizar textos y quitar el tono de IA." }, "fr": { "name": "legal-writing-deai-polish", "summary": "Détecte et supprime les formulations de style IA dans les articles ; utilisé pour la retouche rédactionnelle, l'optimisation de texte et la suppression du ton IA." }, "ja": { "name": "legal-writing-deai-polish", "summary": "文章中の AI 風の表現パターンを検出して除去します。文章のリファイン、テキスト最適化、AI 口調の解消に使用します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:statute-case-retrieval", "slug": "skill-statute-case-retrieval", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Performs systematic retrieval of statutes and case law based on a three-layer retrieval method, links with the claim-basis paths of the preceding preliminary legal analysis, and outputs two documents: the judge edition 'Statutes, Regulations and Case Compendium' and the lawyer edition 'Regulation an" }, "human": { "zh-CN": { "name": "statute-case-retrieval", "summary": "基于三层检索法进行系统性法律法规与案例检索，与前序\"初步法律分析\"请求权路径联动，输出法官版《法律法规及案例汇编》和律师版《法规案例检索报告》双文档，按总控当前事项记录和复盘台账衔接。当用户发送客户编号并提及\"检索法规案例\"、\"法规汇编\"、\"案例检索\"等指令时触发使用。" }, "en": { "name": "statute-case-retrieval", "summary": "Performs systematic retrieval of statutes and case law based on a three-layer retrieval method, links with the claim-basis paths of the preceding preliminary legal analysis, and outputs two documents: the judge edition 'Statutes, Regulations and Case Compendium' and the lawyer edition 'Regulation and Case Retrieval Report', connected with the controller's current-matter record and review ledger. Triggered when the user sends a client number and mentions commands such as 'retrieve statutes and cases', 'statute compendium', or 'case retrieval'." }, "es": { "name": "statute-case-retrieval", "summary": "Realiza una búsqueda sistemática de leyes, reglamentos y jurisprudencia basada en un método de recuperación de tres capas, se enlaza con las rutas de la base de la pretensión del análisis legal preliminar previo y produce dos documentos: la edición para jueces del 'Compendio de leyes, reglamentos y casos' y la edición para abogados del 'Informe de búsqueda de normativa y casos', conectados con el registro de asunto en curso del controlador y el libro de revisión. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como 'buscar leyes y casos', 'compendio normativo' o 'búsqueda de casos'." }, "fr": { "name": "statute-case-retrieval", "summary": "Réalise une recherche systématique de lois, règlements et jurisprudence fondée sur une méthode de recherche à trois niveaux, se raccorde aux voies des fondements de la demande de l'analyse juridique préliminaire précédente et produit deux documents : l'édition juge du 'Recueil des lois, règlements et affaires' et l'édition avocat du 'Rapport de recherche de textes et de jurisprudence', en lien avec l'enregistrement de l'affaire en cours du contrôleur et le registre de révision. Déclenché lorsque l'utilisateur envoie un numéro de client et mentionne des commandes telles que 'rechercher lois et affaires', 'recueil de textes' ou 'recherche de jurisprudence'." }, "ja": { "name": "statute-case-retrieval", "summary": "三層検索法に基づき、法令と判例の体系的な検索を行います。前段の「初歩的法律分析」の請求権ルートと連動し、裁判官版『法令・判例集』と弁護士版『法令・判例検索報告書』の 2 文書を出力し、コントローラの現在事項記録およびレビュー台帳と接続します。ユーザーが顧客番号を送信し、「法令・判例の検索」「法令集」「判例検索」などの指示に言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:special-criminal-procedures", "slug": "skill-special-criminal-procedures", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Skill for defense and representation in special criminal procedures, covering the initiation conditions, material review, hearing or trial preparation, document generation, and hand-off to the stage-specific criminal skills for the confiscation-of-illegal-gains procedure, the compulsory medical trea" }, "human": { "zh-CN": { "name": "special-criminal-procedures", "summary": "刑事特殊程序辩护与代理Skill，覆盖违法所得没收程序、精神病人强制医疗程序、再审申诉程序的启动条件、材料审查、听证或开庭准备、文书生成和与各刑事阶段Skill衔接。当用户提及\"违法所得没收\"、\"强制医疗\"、\"精神病鉴定\"、\"再审\"、\"申诉\"、\"生效判决错误\"等关键词时触发使用。" }, "en": { "name": "special-criminal-procedures", "summary": "Skill for defense and representation in special criminal procedures, covering the initiation conditions, material review, hearing or trial preparation, document generation, and hand-off to the stage-specific criminal skills for the confiscation-of-illegal-gains procedure, the compulsory medical treatment procedure for the mentally ill, and the retrial/petition procedure. Triggered when the user mentions keywords such as 'confiscation of illegal gains', 'compulsory medical treatment', 'psychiatric assessment', 'retrial', 'petition', or 'effective judgment errors'." }, "es": { "name": "special-criminal-procedures", "summary": "Skill de defensa y representación en procedimientos penales especiales: cubre las condiciones de inicio, la revisión de materiales, la preparación de audiencias o juicios, la generación de documentos y el enlace con los skills de cada etapa penal para el procedimiento de decomiso de ganancias ilícitas, el procedimiento de tratamiento médico obligatorio para enfermos mentales y el procedimiento de revisión y queja. Se activa cuando el usuario menciona palabras clave como 'decomiso de ganancias ilícitas', 'tratamiento médico obligatorio', 'peritaje psiquiátrico', 'revisión de sentencia', 'queja' o 'errores de sentencia firme'." }, "fr": { "name": "special-criminal-procedures", "summary": "Skill de défense et de représentation dans les procédures pénales spéciales : couvre les conditions de déclenchement, l'examen des pièces, la préparation des audiences ou du procès, la génération d'actes et l'articulation avec les skills de chaque phase pénale, pour la procédure de confiscation des gains illicites, la procédure de soin médical obligatoire des malades mentaux et la procédure de réexamen et de recours. Déclenché lorsque l'utilisateur mentionne des mots-clés tels que 'confiscation des gains illicites', 'soin médical obligatoire', 'expertise psychiatrique', 'réexamen', 'recours' ou 'erreurs d'un jugement définitif'." }, "ja": { "name": "special-criminal-procedures", "summary": "刑事特別手続の弁護・代理用 skill。違法所得没収手続、精神病人強制医療手続、再審・申立て手続それぞれの開始要件、資料審査、聴聞・公判準備、文書生成、各刑事段階 skill との連携を網羅します。「違法所得の没収」「強制医療」「精神鑑定」「再審」「申立て」「確定判決の誤り」などのキーワードにユーザーが言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:employer-labor-compliance", "slug": "skill-employer-labor-compliance", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Company-side labor and employment compliance skill. Used for the employer's day-to-day labor compliance, updates to employee handbooks/internal rules/labor contracts, wage and working-hours Q&A, drafting internal labor policies, and pre-emptive risk assessment before the company dismisses, transfers" }, "human": { "zh-CN": { "name": "employer-labor-compliance", "summary": "公司端劳动用工合规 Skill。用于用人单位日常劳动合规、员工手册/规章制度/劳动合同更新、工资工时问答、内部劳动政策起草，以及公司在解除、调岗、降薪、处罚、员工投诉或仲裁前的风险预判。用户提到公司劳动合规、员工手册、规章制度、薪酬制度、考勤制度、加班费、发薪日、特殊工时、劳动合同修改、内部政策起草、辞退员工风险、员工投诉公司、调岗降薪等场景时触发。" }, "en": { "name": "employer-labor-compliance", "summary": "Company-side labor and employment compliance skill. Used for the employer's day-to-day labor compliance, updates to employee handbooks/internal rules/labor contracts, wage and working-hours Q&A, drafting internal labor policies, and pre-emptive risk assessment before the company dismisses, transfers, cuts pay, disciplines, receives employee complaints, or faces arbitration. Triggered when the user mentions company labor compliance, employee handbooks, internal rules, compensation systems, attendance systems, overtime pay, paydays, special working hours, labor contract amendments, internal policy drafting, dismissal risk, employee complaints against the company, or transfers and pay cuts." }, "es": { "name": "employer-labor-compliance", "summary": "Skill de cumplimiento laboral por el lado de la empresa. Se usa para el cumplimiento laboral diario del empleador, la actualización de manuales de empleado / reglamentos internos / contratos de trabajo, consultas sobre salarios y jornadas, la redacción de políticas laborales internas y la valoración previa de riesgos antes de que la empresa despida, traslade, reduzca salarios, sancione, reciba quejas de empleados o enfrente un arbitraje. Se activa cuando el usuario menciona cumplimiento laboral empresarial, manual de empleado, reglamentos internos, sistema salarial, control de asistencia, pago de horas extra, día de pago, regímenes especiales de jornada, modificación de contratos de trabajo, redacción de políticas internas, riesgo de despido, quejas de empleados contra la empresa, o traslados y reducciones salariales." }, "fr": { "name": "employer-labor-compliance", "summary": "Skill de conformité en matière de travail côté entreprise. Sert à la conformité quotidienne de l'employeur, aux mises à jour du manuel du personnel / du règlement intérieur / des contrats de travail, aux questions-réponses sur salaires et temps de travail, à la rédaction de politiques internes et à l'anticipation des risques avant que l'entreprise ne licencie, ne mute, ne réduise un salaire, ne sanctionne, ne reçoive une plainte d'un salarié ou ne fasse face à un arbitrage. Déclenché lorsque l'utilisateur mentionne la conformité sociale de l'entreprise, le manuel du personnel, le règlement intérieur, la politique de rémunération, le système de pointage, les heures supplémentaires, le jour de paie, les régimes spéciaux d'horaire, la modification du contrat de travail, la rédaction de politiques internes, le risque de licenciement, les plaintes de salariés contre l'entreprise, ou les mutations et baisses de salaire." }, "ja": { "name": "employer-labor-compliance", "summary": "会社側の労働雇用コンプライアンス skill。使用者の日常的な労務コンプライアンス、就業規則・社内規程・労働契約の更新、賃金・労働時間の質問対応、社内労働ポリシーの起草、および会社が解雇、配置転換、減給、懲戒、従業員からの申立て、仲裁に臨む前のリスク予測に使用します。企業の労務コンプライアンス、就業規則、社内規程、給与制度、勤怠制度、残業代、給与支払日、特殊な労働時間制、労働契約の変更、社内ポリシーの起草、従業員解雇リスク、従業員からの会社への申立て、配置転換・減給などのシーンに言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:regulatory-monitoring", "slug": "skill-regulatory-monitoring", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Skill for monitoring key regulatory developments in mainland China and assessing their compliance impact on clients. Used for regulatory developments, new-rule updates, policy changes, industry regulation, tiering of statutes and policies, client compliance gap checks, policy/system revision recomme" }, "human": { "zh-CN": { "name": "regulatory-monitoring", "summary": "中国境内重点监管动态监测与客户合规影响评估 Skill。用于监管动态、新规更新、政策变化、行业监管、法规政策分级、客户合规缺口核查、政策制度修改建议、整改清单、监管合规简报等任务。适用于用户要求监测某行业、某客户、某产品或某主题的法律法规、监管政策、部门规章、规范性文件、征求意见稿、执法动态，并判断新规要求、客户现有合规文件缺口和整改方案。" }, "en": { "name": "regulatory-monitoring", "summary": "Skill for monitoring key regulatory developments in mainland China and assessing their compliance impact on clients. Used for regulatory developments, new-rule updates, policy changes, industry regulation, tiering of statutes and policies, client compliance gap checks, policy/system revision recommendations, remediation checklists, and regulatory compliance briefings. Applicable when the user asks to monitor, for an industry, client, product, or topic, laws and regulations, regulatory policies, departmental rules, normative documents, draft rules for public comment, or enforcement developments, and to determine new-rule requirements, gaps in the client's existing compliance documents, and remediation plans." }, "es": { "name": "regulatory-monitoring", "summary": "Skill para la vigilancia de novedades regulatorias clave en China continental y la evaluación de su impacto en el cumplimiento de los clientes. Se usa para novedades regulatorias, actualizaciones de nuevas normas, cambios de política, regulación sectorial, clasificación por niveles de leyes y políticas, verificación de brechas de cumplimiento del cliente, recomendaciones de modificación de políticas y sistemas, listas de subsanación y boletines de cumplimiento regulatorio. Aplica cuando el usuario pide monitorear, para una industria, cliente, producto o tema, leyes y reglamentos, políticas regulatorias, normas departamentales, documentos normativos, borradores para comentarios públicos o novedades de fiscalización, y determinar los requisitos de las nuevas normas, las brechas de los documentos de cumplimiento existentes del cliente y los planes de subsanación." }, "fr": { "name": "regulatory-monitoring", "summary": "Skill de veille des évolutions réglementaires clés en Chine continentale et d'évaluation de leur impact conformité sur les clients. Sert au suivi des évolutions réglementaires, des mises à jour de nouvelles règles, des changements de politique, de la régulation sectorielle, de la hiérarchisation des textes, du contrôle des lacunes de conformité des clients, des recommandations de révision des politiques et systèmes, des listes de mise en conformité et des notes de conformité réglementaire. Applicable lorsque l'utilisateur demande de surveiller, pour une industrie, un client, un produit ou un sujet, les lois et règlements, politiques réglementaires, règles ministérielles, documents normatifs, projets soumis à consultation publique ou actualités de l'application de la loi, et de déterminer les exigences des nouvelles règles, les lacunes des documents de conformité existants du client et les plans de remédiation." }, "ja": { "name": "regulatory-monitoring", "summary": "中国本土における重要な規制動向を監視し、顧客へのコンプライアンス影響を評価する skill。規制動向、新規則の更新、政策変更、業界規制、法令・政策の重要度分類、顧客のコンプライアンス欠落確認、政策・制度の改定提案、是正チェックリスト、規制コンプライアンス・ブリーフィングなどのタスクに使用します。ユーザーが特定の業界、顧客、製品、またはテーマについて、法令、規制政策、部門規則、規範的文書、パブリックコメント案、執行動向の監視と、新規則の要求事項、顧客の既存コンプライアンス文書の欠落、是正方案の判断を求める場合に適用されます。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:bankruptcy-reconciliation", "slug": "skill-bankruptcy-reconciliation", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Provides full-process support for the bankruptcy composition procedure, covering the complete flow from the composition application, drafting of the agreement, creditor meeting voting, and confirmation of the agreement's validity, through execution supervision and conversion to liquidation. Applies " }, "human": { "zh-CN": { "name": "bankruptcy-reconciliation", "summary": "提供破产和解程序全流程支持，覆盖从和解申请、协议草案编制、债权人会议表决、协议效力确认到执行监督及转清算的完整流程。适用《企业破产法》第95-99条。当用户发送客户编号并提及\"破产和解\"、\"和解协议\"、\"和解程序\"等指令时触发使用。" }, "en": { "name": "bankruptcy-reconciliation", "summary": "Provides full-process support for the bankruptcy composition procedure, covering the complete flow from the composition application, drafting of the agreement, creditor meeting voting, and confirmation of the agreement's validity, through execution supervision and conversion to liquidation. Applies Articles 95-99 of the Enterprise Bankruptcy Law. Triggered when the user sends a client number and mentions commands such as 'bankruptcy composition', 'composition agreement', or 'composition procedure'." }, "es": { "name": "bankruptcy-reconciliation", "summary": "Brinda soporte a todo el proceso del procedimiento de composición concursal: desde la solicitud de composición, la redacción del borrador del acuerdo, la votación en la junta de acreedores y la confirmación de la validez del acuerdo, hasta la supervisión de la ejecución y la conversión a liquidación. Aplica los artículos 95-99 de la Ley de Quiebras de Empresas. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como 'composición concursal', 'acuerdo de composición' o 'procedimiento de composición'." }, "fr": { "name": "bankruptcy-reconciliation", "summary": "Fournit un accompagnement de bout en bout de la procédure de concordat concursal : du dépôt de la demande, à la rédaction du projet d'accord, au vote en assemblée des créanciers et à la confirmation de la validité de l'accord, jusqu'à la supervision de son exécution et la bascule vers la liquidation. Applique les articles 95 à 99 de la Loi sur la faillite des entreprises. Déclenché lorsque l'utilisateur envoie un numéro de client et mentionne des commandes telles que 'concordat concursal', 'accord de concordat' ou 'procédure de concordat'." }, "ja": { "name": "bankruptcy-reconciliation", "summary": "破産和解手続の全プロセスをサポートします。和解申立て、協議案の作成、債権者会議での表決、協議の効力確認から、執行監督および清算への移行までの流れを網羅します。『企業破産法』第95条から第99条に適用します。ユーザーが顧客番号を送信し、「破産和解」「和解協議」「和解手続」などの指示に言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:bankruptcy-document-generator", "slug": "skill-bankruptcy-document-generator", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Generates all kinds of bankruptcy-procedure documents on demand based on the 165 document templates in the 'Model Styles of Legal Documents for Bankruptcy Proceedings' (106 court documents + 59 administrator documents). This skill serves as the unified generation tool for bankruptcy documents and is" }, "human": { "zh-CN": { "name": "bankruptcy-document-generator", "summary": "基于《破产程序法律文书样式》中的165份文书模板（106份法院文书+59份管理人文书），按需生成各类破产程序文书。本Skill作为破产文书的统一生成工具，供其他破产Skill调用。当用户发送客户编号并提及\"破产文书\"/\"生成XX通知\"/\"生成XX报告\"/\"生成XX裁定\"等指令时触发使用。" }, "en": { "name": "bankruptcy-document-generator", "summary": "Generates all kinds of bankruptcy-procedure documents on demand based on the 165 document templates in the 'Model Styles of Legal Documents for Bankruptcy Proceedings' (106 court documents + 59 administrator documents). This skill serves as the unified generation tool for bankruptcy documents and is called by other bankruptcy skills. Triggered when the user sends a client number and mentions commands such as 'bankruptcy documents', 'generate XX notice', 'generate XX report', or 'generate XX ruling'." }, "es": { "name": "bankruptcy-document-generator", "summary": "Genera a demanda todo tipo de documentos del procedimiento concursal a partir de las 165 plantillas de los 'Modelos de documentos jurídicos de los procedimientos concursales' (106 documentos judiciales + 59 documentos del administrador). Este skill actúa como herramienta unificada de generación de documentos concursales y es invocado por otros skills de quiebra. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como 'documentos concursales', 'generar la notificación XX', 'generar el informe XX' o 'generar el auto XX'." }, "fr": { "name": "bankruptcy-document-generator", "summary": "Génère à la demande tout type d'actes de procédure de faillite à partir des 165 modèles des 'Styles types des documents juridiques des procédures de faillite' (106 documents judiciaires + 59 documents de l'administrateur). Ce skill sert d'outil unifié de génération des actes de faillite et est appelé par les autres skills de faillite. Déclenché lorsque l'utilisateur envoie un numéro de client et mentionne des commandes telles que 'documents de faillite', 'générer la notification XX', 'générer le rapport XX' ou 'générer la décision XX'." }, "ja": { "name": "bankruptcy-document-generator", "summary": "『破産手続法律文書様式』の165種類の文書テンプレート（裁判所文書106種＋管理人文書59種）に基づき、各種破産手続文書をオンデマンドで生成します。本 skill は破産文書の統一生成ツールとして、他の破産 skill から呼び出されます。ユーザーが顧客番号を送信し、「破産文書」「XX通知の生成」「XX報告書の生成」「XX決定の生成」などの指示に言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:bankruptcy-legal-analysis", "slug": "skill-bankruptcy-legal-analysis", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Performs systematic legal analysis of bankruptcy cases based on the Enterprise Bankruptcy Law and its judicial interpretations, supporting three role perspectives - administrator, creditor's counsel, and debtor's counsel - and generating a standardized bankruptcy legal analysis report as a Word docu" }, "human": { "zh-CN": { "name": "bankruptcy-legal-analysis", "summary": "基于《企业破产法》及司法解释，进行破产案件的系统性法律分析，支持管理人、债权人代理、债务人代理三种角色视角，生成标准化破产法律分析报告Word文档。当用户发送客户编号并提及\"破产法律分析\"、\"破产分析\"指令时触发使用，或在破产案件流程中自动衔接（BP1→BP2）。" }, "en": { "name": "bankruptcy-legal-analysis", "summary": "Performs systematic legal analysis of bankruptcy cases based on the Enterprise Bankruptcy Law and its judicial interpretations, supporting three role perspectives - administrator, creditor's counsel, and debtor's counsel - and generating a standardized bankruptcy legal analysis report as a Word document. Triggered when the user sends a client number and mentions the commands 'bankruptcy legal analysis' or 'bankruptcy analysis', or automatically chained within the bankruptcy case flow (BP1→BP2)." }, "es": { "name": "bankruptcy-legal-analysis", "summary": "Realiza un análisis jurídico sistemático de casos de quiebra basado en la Ley de Quiebras de Empresas y sus interpretaciones judiciales, con tres perspectivas de rol - administrador, representante de acreedores y representante del deudor - y genera un informe estandarizado de análisis jurídico de quiebra en documento Word. Se activa cuando el usuario envía un número de cliente y menciona las instrucciones 'análisis jurídico de quiebra' o 'análisis de quiebra', o se encadena automáticamente en el flujo del caso de quiebra (BP1→BP2)." }, "fr": { "name": "bankruptcy-legal-analysis", "summary": "Réalise une analyse juridique systématique des dossiers de faillite fondée sur la Loi sur la faillite des entreprises et ses interprétations judiciaires, avec trois perspectives de rôle - administrateur, conseil des créanciers, conseil du débiteur - et génère un rapport normalisé d'analyse juridique de faillite au format Word. Déclenché lorsque l'utilisateur envoie un numéro de client et mentionne les commandes 'analyse juridique de faillite' ou 'analyse de faillite', ou chaîné automatiquement dans le flux du dossier de faillite (BP1→BP2)." }, "ja": { "name": "bankruptcy-legal-analysis", "summary": "『企業破産法』およびその司法解釈に基づき、破産案件の体系的な法律分析を行います。管理人、債権者代理人、債務者代理人の 3 つのロール視点に対応し、標準化された破産法律分析報告書を Word 文書として生成します。ユーザーが顧客番号を送信し、「破産法律分析」「破産分析」という指示に言及した場合に発動するほか、破産案件フロー内で自動接続（BP1→BP2）されます。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:bankruptcy-liquidation", "slug": "skill-bankruptcy-liquidation", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Covers the full bankruptcy liquidation procedure, from the declaration of bankruptcy to the closure of the proceeding. Primarily work for R1 (the administrator), with R2/R3 participating in oversight. Triggered when the user sends a client number and mentions commands such as 'bankruptcy liquidation" }, "human": { "zh-CN": { "name": "bankruptcy-liquidation", "summary": "覆盖破产清算程序全流程，从宣告破产到程序终结。主要为R1（管理人）工作，R2/R3参与监督。当用户发送客户编号并提及\"破产清算\"、\"财产变价\"、\"财产分配\"、\"终结程序\"等指令时触发使用。" }, "en": { "name": "bankruptcy-liquidation", "summary": "Covers the full bankruptcy liquidation procedure, from the declaration of bankruptcy to the closure of the proceeding. Primarily work for R1 (the administrator), with R2/R3 participating in oversight. Triggered when the user sends a client number and mentions commands such as 'bankruptcy liquidation', 'asset monetization', 'asset distribution', or 'close the proceeding'." }, "es": { "name": "bankruptcy-liquidation", "summary": "Cubre el proceso completo del procedimiento de liquidación concursal, desde la declaración de quiebra hasta el cierre del procedimiento. Es principalmente trabajo de R1 (el administrador), con participación de R2/R3 en la supervisión. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como 'liquidación concursal', 'realización de bienes', 'distribución de bienes' o 'cierre del procedimiento'." }, "fr": { "name": "bankruptcy-liquidation", "summary": "Couvre l'intégralité de la procédure de liquidation de faillite, de la déclaration de faillite à la clôture de la procédure. Travail principalement destiné à R1 (l'administrateur), R2/R3 participant à la supervision. Déclenché lorsque l'utilisateur envoie un numéro de client et mentionne des commandes telles que 'liquidation de faillite', 'réalisation des actifs', 'répartition des actifs' ou 'clôture de la procédure'." }, "ja": { "name": "bankruptcy-liquidation", "summary": "破産清算手続の全プロセスを網羅します。破産宣告から手続終結までを対象とし、主に R1（管理人）の業務で、R2/R3 が監督に参加します。ユーザーが顧客番号を送信し、「破産清算」「財産の換価」「財産分配」「手続終結」などの指示に言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:bankruptcy-filing-acceptance", "slug": "skill-bankruptcy-filing-acceptance", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Covers the full bankruptcy application process, including drafting the petition, preparing attachments, submission tracking, and initial work after acceptance. Triggered when the user sends a client number and mentions commands such as 'bankruptcy application', 'reorganization application', or 'comp" }, "human": { "zh-CN": { "name": "bankruptcy-filing-acceptance", "summary": "覆盖破产申请全流程，包括申请书起草、附件准备、提交跟踪、受理后初步工作。当用户发送客户编号并提及\"破产申请\"、\"重整申请\"、\"和解申请\"等指令时触发使用。" }, "en": { "name": "bankruptcy-filing-acceptance", "summary": "Covers the full bankruptcy application process, including drafting the petition, preparing attachments, submission tracking, and initial work after acceptance. Triggered when the user sends a client number and mentions commands such as 'bankruptcy application', 'reorganization application', or 'composition application'." }, "es": { "name": "bankruptcy-filing-acceptance", "summary": "Cubre el proceso completo de la solicitud concursal: redacción de la solicitud, preparación de anexos, seguimiento de la presentación y trabajos iniciales tras la admisión. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como 'solicitud de quiebra', 'solicitud de reorganización' o 'solicitud de composición'." }, "fr": { "name": "bankruptcy-filing-acceptance", "summary": "Couvre l'ensemble du processus de demande de faillite : rédaction de la requête, préparation des annexes, suivi du dépôt et travaux préliminaires après acceptation. Déclenché lorsque l'utilisateur envoie un numéro de client et mentionne des commandes telles que 'demande de faillite', 'demande de redressement' ou 'demande de concordat'." }, "ja": { "name": "bankruptcy-filing-acceptance", "summary": "破産申立ての全プロセスを網羅します。申立書の起草、添付資料の準備、提出追跡、受理後の初期作業を含みます。ユーザーが顧客番号を送信し、「破産申立て」「重整申立て」「和解申立て」などの指示に言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:case-filing-management", "slug": "skill-case-filing-management", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Systematically guides the full first-instance civil case-filing process, including filing material checks, jurisdiction pre-checks, filing method selection, court fee calculation, and evidence deadline management. Triggered when the user sends a client number and mentions commands such as 'file the " }, "human": { "zh-CN": { "name": "case-filing-management", "summary": "系统化指引民事一审立案全流程，包括立案材料核查、管辖权预检、立案方式选择、诉讼费计算、举证期限管理等。当用户发送客户编号并提及\"立案\"、\"去立案\"、\"立案准备\"、\"立案材料\"、\"网上立案\"等指令时触发使用。" }, "en": { "name": "case-filing-management", "summary": "Systematically guides the full first-instance civil case-filing process, including filing material checks, jurisdiction pre-checks, filing method selection, court fee calculation, and evidence deadline management. Triggered when the user sends a client number and mentions commands such as 'file the case', 'go file', 'filing preparation', 'filing materials', or 'online filing'." }, "es": { "name": "case-filing-management", "summary": "Orienta de manera sistemática el proceso completo de presentación de la demanda en primera instancia civil: verificación de los materiales de presentación, prechequeo de jurisdicción y competencia, elección de la vía de presentación, cálculo de las tasas judiciales y gestión de los plazos probatorios. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como 'presentar la demanda', 'ir a presentar', 'preparación de la presentación', 'materiales de presentación' o 'presentación en línea'." }, "fr": { "name": "case-filing-management", "summary": "Guide de manière systématique l'ensemble du processus de saisine en première instance civile : vérification des pièces de saisine, précontrôle de compétence, choix du mode de dépôt, calcul des frais de justice et gestion des délais de preuve. Déclenché lorsque l'utilisateur envoie un numéro de client et mentionne des commandes telles que 'déposer la demande', 'aller déposer', 'préparation du dépôt', 'pièces de saisine' ou 'dépôt en ligne'." }, "ja": { "name": "case-filing-management", "summary": "民事第一審の立案（訴訟提起）プロセス全体を体系的にガイドします。立案資料のチェック、管轄権の事前確認、立案方法の選択、訴訟費用の計算、立証期限の管理などを含みます。ユーザーが顧客番号を送信し、「立案」「立案しに行く」「立案準備」「立案資料」「オンライン立案」などの指示に言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:simplified-summary-procedure", "slug": "skill-simplified-summary-procedure", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Branch skill for criminal defense in the simplified procedure, the expedited procedure, and streamlined trial under the ordinary procedure, focusing on procedural-applicability review, sentencing negotiation, trial simplification, rights protection, and time-node management in guilty-plea cases. As " }, "human": { "zh-CN": { "name": "simplified-summary-procedure", "summary": "刑事辩护简易程序、速裁程序及普通程序简化审理分支Skill，重点处理认罪认罚案件中的程序适用审查、量刑协商、庭审简化、权利保障和时间节点管理。作为一审阶段辩护的分支，当用户提及\"简易程序\"、\"速裁程序\"、\"认罪认罚\"、\"快速审理\"、\"当庭宣判\"等关键词时触发使用。" }, "en": { "name": "simplified-summary-procedure", "summary": "Branch skill for criminal defense in the simplified procedure, the expedited procedure, and streamlined trial under the ordinary procedure, focusing on procedural-applicability review, sentencing negotiation, trial simplification, rights protection, and time-node management in guilty-plea cases. As a branch of first-instance defense work, it is triggered when the user mentions keywords such as 'simplified procedure', 'expedited procedure', 'guilty plea and acceptance of punishment', 'expedited trial', or 'same-day judgment'." }, "es": { "name": "simplified-summary-procedure", "summary": "Skill de rama para la defensa penal en el procedimiento simplificado, el procedimiento de resolución rápida y el juicio abreviado del procedimiento ordinario; se centra en la revisión de la procedencia del procedimiento, la negociación de la pena, la simplificación del juicio, la garantía de derechos y la gestión de hitos temporales en casos de confesión y aceptación de sanción. Como rama de la defensa en primera instancia, se activa cuando el usuario menciona palabras clave como 'procedimiento simplificado', 'procedimiento de resolución rápida', 'confesión y aceptación de sanción', 'juicio rápido' o 'sentencia inmediata'." }, "fr": { "name": "simplified-summary-procedure", "summary": "Skill de branche pour la défense pénale dans la procédure simplifiée, la procédure accélérée et le jugement abrégé au sein de la procédure ordinaire ; il traite en priorité l'examen de l'applicabilité procédurale, la négociation sur la peine, la simplification des débats, la protection des droits et la gestion des jalons temporels dans les affaires avec reconnaissance de culpabilité et acceptation de la peine. En tant que branche de la défense en première instance, il se déclenche lorsque l'utilisateur mentionne des mots-clés tels que 'procédure simplifiée', 'procédure accélérée', 'reconnaissance de culpabilité et acceptation de la peine', 'jugement rapide' ou 'jugement immédiat'." }, "ja": { "name": "simplified-summary-procedure", "summary": "刑事弁護における簡易手続、速裁手続、通常手続の簡略審理に対応する分岐 skill。認罪認罰（罪を認めて刑罰を受け入れる）事件における手続適用の審査、量刑交渉、公判の簡略化、権利保障、時間ノード管理を重点的に扱います。第一審弁護の分岐として、「簡易手続」「速裁手続」「認罪認罰」「迅速審理」「即日判決」などのキーワードにユーザーが言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:bankruptcy-administrator-tasks", "slug": "skill-bankruptcy-administrator-tasks", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Skill exclusive to R1 (the administrator), covering all initial post-acceptance duties of the administrator, including seal and account management, taking over assets, internal affairs management, contract handling, hiring personnel, administrator team management, and remuneration plan design. Trigg" }, "human": { "zh-CN": { "name": "bankruptcy-administrator-tasks", "summary": "R1（管理人）专属Skill，覆盖管理人受理后的全部初期工作职责，包括印章与账户管理、财产接管、内部事务管理、合同处理、聘用工作人员、管理人团队管理及报酬方案制定。当用户发送客户编号并提及\"管理人接管\"、\"开始接管\"、\"接管工作\"等指令时触发使用。" }, "en": { "name": "bankruptcy-administrator-tasks", "summary": "Skill exclusive to R1 (the administrator), covering all initial post-acceptance duties of the administrator, including seal and account management, taking over assets, internal affairs management, contract handling, hiring personnel, administrator team management, and remuneration plan design. Triggered when the user sends a client number and mentions commands such as 'administrator takeover', 'start the takeover', or 'takeover work'." }, "es": { "name": "bankruptcy-administrator-tasks", "summary": "Skill exclusivo de R1 (el administrador): cubre todas las funciones iniciales del administrador tras la admisión, incluidas la gestión de sellos y cuentas, la toma de posesión de los bienes, la gestión de asuntos internos, el tratamiento de contratos, la contratación de personal, la gestión del equipo del administrador y la elaboración del plan de retribución. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como 'toma de posesión del administrador', 'iniciar la toma de posesión' o 'trabajos de toma de posesión'." }, "fr": { "name": "bankruptcy-administrator-tasks", "summary": "Skill exclusif à R1 (l'administrateur) : couvre l'ensemble des fonctions initiales de l'administrateur après acceptation, notamment la gestion des sceaux et des comptes, la prise en charge des actifs, la gestion des affaires internes, le traitement des contrats, le recrutement de personnel, la gestion de l'équipe de l'administrateur et l'élaboration du barème de rémunération. Déclenché lorsque l'utilisateur envoie un numéro de client et mentionne des commandes telles que 'prise en charge par l'administrateur', 'démarrer la prise en charge' ou 'travaux de prise en charge'." }, "ja": { "name": "bankruptcy-administrator-tasks", "summary": "R1（管理人）専用 skill。受理後の管理人の初期業務をすべて網羅し、印章・口座管理、財産の引き受け、内部事務の管理、契約の処理、職員の雇用、管理人チームの管理、報酬案の策定を含みます。ユーザーが顧客番号を送信し、「管理人の引き受け」「引き受け開始」「引き受け業務」などの指示に言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:case-closure-archive", "slug": "skill-case-closure-archive", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "After judgment/mediation is completed, systematically handles case-closure work including case file archiving, case summary, client hand-over, and fee settlement. Triggered when the user sends a client number and mentions commands such as 'archive', 'close the case', 'case summary', 'case-closing me" }, "human": { "zh-CN": { "name": "case-closure-archive", "summary": "在判决/调解完成后，系统化完成案卷归档、案件总结、客户移交、费用结算等结案工作。当用户发送客户编号并提及\"归档\"、\"结案\"、\"案件总结\"、\"办案小结\"、\"案卷移交\"、\"费用结算\"等指令时触发使用。" }, "en": { "name": "case-closure-archive", "summary": "After judgment/mediation is completed, systematically handles case-closure work including case file archiving, case summary, client hand-over, and fee settlement. Triggered when the user sends a client number and mentions commands such as 'archive', 'close the case', 'case summary', 'case-closing memo', 'file hand-over', or 'fee settlement'." }, "es": { "name": "case-closure-archive", "summary": "Una vez concluido el fallo o la mediación, lleva a cabo de forma sistemática las tareas de cierre: archivo del expediente, resumen del caso, traspaso al cliente y liquidación de honorarios. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como 'archivar', 'cerrar el caso', 'resumen del caso', 'memoria de cierre', 'traspaso del expediente' o 'liquidación de honorarios'." }, "fr": { "name": "case-closure-archive", "summary": "Une fois le jugement ou la médiation achevé, mène de façon systématique les travaux de clôture : archivage du dossier, synthèse de l'affaire, passation au client et règlement des honoraires. Déclenché lorsque l'utilisateur envoie un numéro de client et mentionne des commandes telles que 'archiver', 'clôturer le dossier', 'synthèse de l'affaire', 'note de clôture', 'passation du dossier' ou 'règlement des honoraires'." }, "ja": { "name": "case-closure-archive", "summary": "判決・調停の完了後、事件ファイルのアーカイブ、案件総括、顧客への引き継ぎ、費用精算などの結案（クロージング）作業を体系的に行います。ユーザーが顧客番号を送信し、「アーカイブ」「結案」「案件総括」「処理報告」「ファイル引き継ぎ」「費用精算」などの指示に言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:litigation-visualization", "slug": "skill-litigation-visualization", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Generates litigation visualization charts in Mermaid syntax (legal-relationship diagram, timeline, issue tree, element analysis diagram, offense-defense confrontation diagram, evidence-chain graph), connects with the controller's current-matter record and review ledger, and produces a litigation vis" }, "human": { "zh-CN": { "name": "litigation-visualization", "summary": "基于Mermaid语法生成诉讼可视化图表（法律关系图、时间轴图、争点树、要件分析图、攻防对抗图、证据链图谱），按总控当前事项记录和复盘台账衔接，生成诉讼可视化报告Word文档（法官版+律师版）。当用户发送客户编号并提及\"可视化\"、\"诉讼可视化\"、\"制作图表\"等指令时触发使用。" }, "en": { "name": "litigation-visualization", "summary": "Generates litigation visualization charts in Mermaid syntax (legal-relationship diagram, timeline, issue tree, element analysis diagram, offense-defense confrontation diagram, evidence-chain graph), connects with the controller's current-matter record and review ledger, and produces a litigation visualization report as a Word document (judge edition + lawyer edition). Triggered when the user sends a client number and mentions commands such as 'visualization', 'litigation visualization', or 'make charts'." }, "es": { "name": "litigation-visualization", "summary": "Genera gráficos de visualización de litigios en sintaxis Mermaid (diagrama de relaciones jurídicas, línea de tiempo, árbol de cuestiones, diagrama de análisis de elementos, diagrama de confrontación ataque-defensa, grafo de cadena probatoria), se conecta con el registro de asunto en curso del controlador y el libro de revisión, y produce un informe de visualización de litigios en documento Word (edición juez + edición abogado). Se activa cuando el usuario envía un número de cliente y menciona instrucciones como 'visualización', 'visualización de litigios' o 'elaborar gráficos'." }, "fr": { "name": "litigation-visualization", "summary": "Génère des schémas de visualisation du contentieux en syntaxe Mermaid (diagramme des relations juridiques, frise chronologique, arbre des points en litige, diagramme d'analyse des éléments, diagramme d'affrontement attaque-défense, graphe de la chaîne de preuves), se raccorde à l'enregistrement de l'affaire en cours du contrôleur et au registre de révision, et produit un rapport de visualisation contentieuse au format Word (édition juge + édition avocat). Déclenché lorsque l'utilisateur envoie un numéro de client et mentionne des commandes telles que 'visualisation', 'visualisation du contentieux' ou 'créer des schémas'." }, "ja": { "name": "litigation-visualization", "summary": "Mermaid 構文で訴訟ビジュアライゼーション図（法律関係図、タイムライン図、争点ツリー、要件分析図、攻防対峙図、証拠チェーングラフ）を生成し、コントローラの現在事項記録およびレビュー台帳と接続して、訴訟ビジュアライゼーション報告書を Word 文書（裁判官版＋弁護士版）として出力します。ユーザーが顧客番号を送信し、「ビジュアライゼーション」「訴訟の可視化」「図表の作成」などの指示に言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:litigation-document-drafting", "slug": "skill-litigation-document-drafting", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Drafts litigation documents such as complaints, statements of defense, attorney arguments, cross-examination opinions, and preservation applications based on the 'six-source system' and claim-basis analysis. Integrates the format standards of the 'Compendium of Legal Documents' (the Zhengzhou Lawyer" }, "human": { "zh-CN": { "name": "litigation-document-drafting", "summary": "基于\"六来源体系\"和请求权基础分析，起草起诉状、答辩状、代理词、质证意见、保全申请书等诉讼文书。整合《法律文书汇编》（郑州市律师协会2023年度优秀法律文书汇编）格式标准，按总控当前事项记录和复盘台账衔接。当用户发送客户编号并提及\"起诉状\"、\"答辩状\"、\"代理词\"、\"质证\"、\"保全\"等指令时触发使用。" }, "en": { "name": "litigation-document-drafting", "summary": "Drafts litigation documents such as complaints, statements of defense, attorney arguments, cross-examination opinions, and preservation applications based on the 'six-source system' and claim-basis analysis. Integrates the format standards of the 'Compendium of Legal Documents' (the Zhengzhou Lawyers Association 2023 compendium of outstanding legal documents) and connects with the controller's current-matter record and review ledger. Triggered when the user sends a client number and mentions commands such as 'complaint', 'statement of defense', 'attorney argument', 'cross-examination', or 'preservation'." }, "es": { "name": "litigation-document-drafting", "summary": "Redacta documentos procesales como demandas, contestaciones, alegatos de la parte, observaciones de contrainterrogatorio y solicitudes cautelares, basándose en el 'sistema de seis fuentes' y el análisis de las bases de la pretensión. Integra los estándares de formato del 'Compendio de documentos jurídicos' (compendio 2023 de documentos jurídicos destacados del Colegio de Abogados de Zhengzhou) y se conecta con el registro de asunto en curso del controlador y el libro de revisión. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como 'demanda', 'contestación', 'alegato', 'contrainterrogatorio' o 'medida cautelar'." }, "fr": { "name": "litigation-document-drafting", "summary": "Rédige des actes de procédure tels que requêtes introductives, conclusions en défense, plaidoiries écrites, observations sur le contre-interrogatoire et requêtes conservatoires, en s'appuyant sur le 'système des six sources' et l'analyse des fondements de la demande. Intègre les standards de format du 'Recueil de documents juridiques' (recueil 2023 des meilleurs documents juridiques de l'Association des avocats de Zhengzhou) et se raccorde à l'enregistrement de l'affaire en cours du contrôleur et au registre de révision. Déclenché lorsque l'utilisateur envoie un numéro de client et mentionne des commandes telles que 'requête introductive', 'conclusions en défense', 'plaidoirie', 'contre-interrogatoire' ou 'mesure conservatoire'." }, "ja": { "name": "litigation-document-drafting", "summary": "「六ソース体系」と請求権基礎分析に基づき、訴状、答弁状、代理人意見書、反対尋問意見、保全申立て書などの訴訟文書を起草します。『法律文書集』（鄭州市弁護士協会 2023 年度優秀法律文書集）のフォーマット標準を統合し、コントローラの現在事項記録およびレビュー台帳と接続します。ユーザーが顧客番号を送信し、「訴状」「答弁状」「代理人意見書」「反対尋問」「保全」などの指示に言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:litigation-case-management", "slug": "skill-litigation-case-management", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Case operations management skill shared by civil, criminal, and labor dispute litigation. Used for case updates, handling of summonses/trial notices/evidence notices/response notices/judgment documents, deadline ledgers, Feishu calendar reminders, maintenance of fact and procedural timelines, Feishu" }, "human": { "zh-CN": { "name": "litigation-case-management", "summary": "民事、刑事及劳动争议诉讼共用的案件运营管理 Skill。用于案件更新、传票/开庭通知/举证通知/应诉通知/裁判文书处理、期限台账、飞书日历提醒、事实时间线和程序时间线维护、飞书思维导图同步、案件简报、诉讼案件总览、案件组合状态、案件关闭与结案归档检查。案件隔离、建档、材料读取复查、来源披露、OCR 校正、缺口归档和法规核验由法律工作总控统一处理。" }, "en": { "name": "litigation-case-management", "summary": "Case operations management skill shared by civil, criminal, and labor dispute litigation. Used for case updates, handling of summonses/trial notices/evidence notices/response notices/judgment documents, deadline ledgers, Feishu calendar reminders, maintenance of fact and procedural timelines, Feishu mind-map sync, case briefings, litigation case overviews, case portfolio status, and case closing with closure-archive checks. Case isolation, file creation, file-read re-checking, source disclosure, OCR correction, gap archiving, and statute verification are handled uniformly by the legal work controller." }, "es": { "name": "litigation-case-management", "summary": "Skill de gestión operativa de casos compartido por litigios civiles, penales y conflictos laborales. Se usa para actualizaciones del caso, tramitación de citaciones / notificaciones de audiencia / notificaciones probatorias / notificaciones de contestación / resoluciones, libro de plazos, recordatorios en el calendario de Feishu, mantenimiento de líneas temporales de hechos y del procedimiento, sincronización de mapas mentales en Feishu, informes de caso, vista general de litigios, estado del portafolio de casos y cierre con verificación de archivo. El aislamiento de casos, la creación de expedientes, la reverificación de lecturas, la divulgación de fuentes, la corrección OCR, el archivo de brechas y la validación de leyes los gestiona de forma unificada el controlador del trabajo legal." }, "fr": { "name": "litigation-case-management", "summary": "Skill de gestion opérationnelle des dossiers partagé par les contentieux civils, pénaux et les litiges du travail. Sert aux mises à jour du dossier, au traitement des assignations / convocations / notifications de preuve / notifications de défense / décisions, au registre des délais, aux rappels du calendrier Feishu, à la tenue des frises de faits et de procédure, à la synchronisation des cartes mentales Feishu, aux notes de dossier, à la vue d'ensemble du contentieux, à l'état du portefeuille de dossiers et à la clôture avec contrôles d'archivage. L'isolation des dossiers, l'ouverture des dossiers, la re-vérification des lectures, la divulgation des sources, la correction OCR, l'archivage des lacunes et la validation des lois sont gérés de façon unifiée par le contrôleur du travail juridique." }, "ja": { "name": "litigation-case-management", "summary": "民事・刑事・労務紛争訴訟で共用する案件オペレーション管理 skill。案件更新、呼出状・開廷通知・立証通知・応訴通知・裁判文書の処理、期限台帳、Feishu カレンダーのリマインダー、事実タイムラインと手続タイムラインの維持、Feishu マインドマップの同期、案件ブリーフィング、訴訟案件の一覧、案件ポートフォリオ状況、案件クローズと結案アーカイブ確認に使用します。案件隔離、ファイル作成、資料読み込み再確認、出典開示、OCR 補正、欠落アーカイブ、法令検証は法律業務コントローラが統一的に処理します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:evidence-collection-management", "slug": "skill-evidence-collection-management", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Assesses evidence gaps based on claim-element analysis and systematically guides the full process of evidence collection, investigative evidence-gathering, witness management, and asset preservation, producing evidence plans, various application documents, and investigation transcripts. Triggered wh" }, "human": { "zh-CN": { "name": "evidence-collection-management", "summary": "基于请求权要件分析进行证据缺口评估，系统化指引证据收集、调查取证、证人管理和财产保全全流程，生成取证方案、各类申请文书和调查笔录。当用户发送客户编号并提及\"调查取证\"、\"证据收集\"、\"财产保全\"、\"调查令\"、\"证人出庭\"、\"补充材料\"、\"取证方案\"等指令时触发使用。" }, "en": { "name": "evidence-collection-management", "summary": "Assesses evidence gaps based on claim-element analysis and systematically guides the full process of evidence collection, investigative evidence-gathering, witness management, and asset preservation, producing evidence plans, various application documents, and investigation transcripts. Triggered when the user sends a client number and mentions commands such as 'investigation and evidence gathering', 'evidence collection', 'asset preservation', 'investigation order', 'witness appearance', 'supplementary materials', or 'evidence plan'." }, "es": { "name": "evidence-collection-management", "summary": "Evalúa las brechas probatorias basándose en el análisis de los elementos de la pretensión y orienta sistemáticamente el proceso completo de recopilación de pruebas, obtención de pruebas por investigación, gestión de testigos y conservación de bienes, generando planes probatorios, diversos escritos de solicitud y actas de investigación. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como 'investigación y obtención de pruebas', 'recopilación de pruebas', 'conservación de bienes', 'orden de investigación', 'comparecencia de testigos', 'documentación complementaria' o 'plan de pruebas'." }, "fr": { "name": "evidence-collection-management", "summary": "Évalue les lacunes probatoires à partir de l'analyse des éléments de la demande et guide systématiquement l'ensemble du processus de collecte de preuves, d'enquête probatoire, de gestion des témoins et de mesures conservatoires sur les actifs, en produisant des plans de preuve, diverses requêtes et procès-verbaux d'enquête. Déclenché lorsque l'utilisateur envoie un numéro de client et mentionne des commandes telles que 'enquête et obtention de preuves', 'collecte de preuves', 'mesures conservatoires sur les actifs', 'ordre d'investigation', 'comparution des témoins', 'pièces complémentaires' ou 'plan de preuve'." }, "ja": { "name": "evidence-collection-management", "summary": "請求権要件分析に基づいて証拠の欠落を評価し、証拠収集、調査取証、証人管理、財産保全の全プロセスを体系的にガイドし、取証プラン、各種申立て文書、調査調書を生成します。ユーザーが顧客番号を送信し、「調査取証」「証拠収集」「財産保全」「調査令状」「証人出廷」「補足資料」「取証プラン」などの指示に言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:mediation-and-settlement", "slug": "skill-mediation-and-settlement", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Provides full-process support for mediation and settlement, including feasibility analysis, drafting mediation proposals, review of mediation statements/settlement agreements, and withdrawal-of-claim management. Triggered when the user sends a client number and mentions commands such as 'mediation'," }, "human": { "zh-CN": { "name": "mediation-and-settlement", "summary": "提供调解与和解全流程支持，包括可行性分析、调解方案起草、调解书/和解协议审查、撤诉管理等功能。当用户发送客户编号并提及\"调解\"、\"和解\"、\"撤诉\"、\"调解方案\"、\"调解书\"等指令时触发使用。" }, "en": { "name": "mediation-and-settlement", "summary": "Provides full-process support for mediation and settlement, including feasibility analysis, drafting mediation proposals, review of mediation statements/settlement agreements, and withdrawal-of-claim management. Triggered when the user sends a client number and mentions commands such as 'mediation', 'settlement', 'withdraw the claim', 'mediation proposal', or 'mediation statement'." }, "es": { "name": "mediation-and-settlement", "summary": "Brinda soporte a todo el proceso de mediación y transacción: análisis de viabilidad, redacción de propuestas de mediación, revisión de actas de mediación / acuerdos transaccionales y gestión de la retirada de la demanda. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como 'mediación', 'transacción', 'retirar la demanda', 'propuesta de mediación' o 'acta de mediación'." }, "fr": { "name": "mediation-and-settlement", "summary": "Fournit un accompagnement de bout en bout pour la médiation et la transaction : analyse de faisabilité, rédaction de propositions de médiation, examen des comptes rendus de médiation / accords transactionnels et gestion du désistement. Déclenché lorsque l'utilisateur envoie un numéro de client et mentionne des commandes telles que 'médiation', 'transaction', 'désistement', 'proposition de médiation' ou 'compte rendu de médiation'." }, "ja": { "name": "mediation-and-settlement", "summary": "調停・和解の全プロセスをサポートします。実現可能性分析、調停案の起草、調停調書・和解契約の審査、訴え取下げの管理などの機能を含みます。ユーザーが顧客番号を送信し、「調停」「和解」「訴えの取下げ」「調停案」「調停調書」などの指示に言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:asset-investigation-management", "slug": "skill-asset-investigation-management", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Skill exclusive to R1 (the administrator), covering comprehensive asset investigation in bankruptcy proceedings, recovery of voidable/invalid acts, and preparation of asset management plans. Triggered when the user sends a client number and mentions commands such as 'asset investigation', 'asset inv" }, "human": { "zh-CN": { "name": "asset-investigation-management", "summary": "R1（管理人）专属Skill，覆盖破产程序中的财产全面调查、可撤销/无效行为追回、财产管理方案编制。当用户发送客户编号并提及\"财产调查\"、\"财产清查\"、\"可撤销行为\"、\"追回财产\"、\"财产管理方案\"等指令时触发使用。" }, "en": { "name": "asset-investigation-management", "summary": "Skill exclusive to R1 (the administrator), covering comprehensive asset investigation in bankruptcy proceedings, recovery of voidable/invalid acts, and preparation of asset management plans. Triggered when the user sends a client number and mentions commands such as 'asset investigation', 'asset inventory', 'voidable acts', 'asset recovery', or 'asset management plan'." }, "es": { "name": "asset-investigation-management", "summary": "Skill exclusivo de R1 (el administrador): cubre la investigación integral de bienes en el procedimiento concursal, la recuperación de actos anulables o nulos y la elaboración de planes de gestión de bienes. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como 'investigación de bienes', 'inventario de bienes', 'actos anulables', 'recuperación de bienes' o 'plan de gestión de bienes'." }, "fr": { "name": "asset-investigation-management", "summary": "Skill exclusif à R1 (l'administrateur) : couvre l'enquête patrimoniale complète dans la procédure de faillite, la reprise des actes annulables ou nuls et l'élaboration des plans de gestion des actifs. Déclenché lorsque l'utilisateur envoie un numéro de client et mentionne des commandes telles que 'enquête patrimoniale', 'inventaire des actifs', 'actes annulables', 'récupération des actifs' ou 'plan de gestion des actifs'." }, "ja": { "name": "asset-investigation-management", "summary": "R1（管理人）専用 skill。破産手続における財産の全面調査、取消可能・無効行為の取り戻し、財産管理方案の作成を網羅します。ユーザーが顧客番号を送信し、「財産調査」「財産の棚卸」「取消可能行為」「財産の取り戻し」「財産管理方案」などの指示に言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:bankruptcy-reorganization", "slug": "skill-bankruptcy-reorganization", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Full-process management of the bankruptcy reorganization procedure, covering all stages: reorganization application and acceptance, management during the reorganization period, drafting the reorganization plan, class voting by creditors, court compulsory approval, and execution and supervision of th" }, "human": { "zh-CN": { "name": "bankruptcy-reorganization", "summary": "破产重整程序全流程管理。覆盖重整申请与受理、重整期间管理、重整计划草案编制、债权分组表决、法院强制批准、重整计划执行与监督等全部环节。适用于管理人（R1）、债权人代理（R2）、债务人代理（R3）三类角色。当用户发送客户编号并提及\"重整\"、\"重整计划\"、\"重整程序\"等指令时触发使用。" }, "en": { "name": "bankruptcy-reorganization", "summary": "Full-process management of the bankruptcy reorganization procedure, covering all stages: reorganization application and acceptance, management during the reorganization period, drafting the reorganization plan, class voting by creditors, court compulsory approval, and execution and supervision of the reorganization plan. Applicable to the three roles of administrator (R1), creditor's counsel (R2), and debtor's counsel (R3). Triggered when the user sends a client number and mentions commands such as 'reorganization', 'reorganization plan', or 'reorganization procedure'." }, "es": { "name": "bankruptcy-reorganization", "summary": "Gestión de proceso completo del procedimiento de reorganización concursal: cubre todas las etapas, desde la solicitud y admisión de la reorganización, la gestión durante el período de reorganización, la redacción del borrador del plan de reorganización, la votación por clases de acreedores y la aprobación forzosa del tribunal, hasta la ejecución y supervisión del plan. Aplica a los tres roles de administrador (R1), representante de acreedores (R2) y representante del deudor (R3). Se activa cuando el usuario envía un número de cliente y menciona instrucciones como 'reorganización', 'plan de reorganización' o 'procedimiento de reorganización'." }, "fr": { "name": "bankruptcy-reorganization", "summary": "Gestion de bout en bout de la procédure de redressement dans la faillite : couvre toutes les étapes, de la demande et l'acceptation du redressement, la gestion pendant la période de redressement, la rédaction du projet de plan de redressement, le vote par classes de créanciers et l'approbation forcée du tribunal, jusqu'à l'exécution et la supervision du plan. Applicable aux trois rôles d'administrateur (R1), de conseil des créanciers (R2) et de conseil du débiteur (R3). Déclenché lorsque l'utilisateur envoie un numéro de client et mentionne des commandes telles que 'redressement', 'plan de redressement' ou 'procédure de redressement'." }, "ja": { "name": "bankruptcy-reorganization", "summary": "破産重整手続の全プロセス管理。重整申立てと受理、重整期間の管理、重整計画案の作成、債権者グループ別表決、裁判所による強制認可、重整計画の執行と監督など全段階を網羅します。管理人（R1）、債権者代理人（R2）、債務者代理人（R3）の 3 ロールに対応します。ユーザーが顧客番号を送信し、「重整」「重整計画」「重整手続」などの指示に言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:food-label-compliance", "slug": "skill-food-label-compliance", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Prepackaged food label compliance review skill, used to review whether food labels conform to GB 7718 (General Standard for the Labeling of Prepackaged Foods) and GB 28050 (General Standard for Nutrition Labeling of Prepackaged Foods). Applicable when: (1) the user submits food label images, text, o" }, "human": { "zh-CN": { "name": "food-label-compliance", "summary": "预包装食品标签合规审核技能，用于审核食品标签是否符合 GB 7718（预包装食品标签通则）和 GB 28050（预包装食品营养标签通则）。适用场景：(1) 用户提交食品标签图片、文字或文档要求合规审核时；(2) 用户提到\"食品标签审核\"\"标签合规\"\"营养标签审查\"\"GB 7718\"\"GB 28050\"等关键词时；(3) 用户要求检查食品标签是否存在缺项、错误或违规风险时。支持 2011 版和 2025 版标准。" }, "en": { "name": "food-label-compliance", "summary": "Prepackaged food label compliance review skill, used to review whether food labels conform to GB 7718 (General Standard for the Labeling of Prepackaged Foods) and GB 28050 (General Standard for Nutrition Labeling of Prepackaged Foods). Applicable when: (1) the user submits food label images, text, or documents for compliance review; (2) the user mentions keywords such as 'food label review', 'label compliance', 'nutrition label review', 'GB 7718', or 'GB 28050'; (3) the user asks to check food labels for missing items, errors, or violation risks. Supports both the 2011 and 2025 versions of the standards." }, "es": { "name": "food-label-compliance", "summary": "Skill de revisión de cumplimiento de etiquetas de alimentos preenvasados, para verificar si las etiquetas se ajustan a la GB 7718 (Norma general de etiquetado de alimentos preenvasados) y la GB 28050 (Norma general de etiquetado nutricional de alimentos preenvasados). Escenarios de aplicación: (1) el usuario envía imágenes, texto o documentos de etiquetas de alimentos para revisión de cumplimiento; (2) el usuario menciona palabras clave como 'revisión de etiquetas de alimentos', 'cumplimiento de etiquetas', 'revisión de etiquetado nutricional', 'GB 7718' o 'GB 28050'; (3) el usuario pide comprobar si la etiqueta presenta omisiones, errores o riesgos de infracción. Admite las versiones 2011 y 2025 de las normas." }, "fr": { "name": "food-label-compliance", "summary": "Skill de contrôle de conformité des étiquettes de denrées alimentaires préemballées, pour vérifier la conformité des étiquettes à la GB 7718 (norme générale d'étiquetage des denrées préemballées) et à la GB 28050 (norme générale d'étiquetage nutritionnel des denrées préemballées). Scénarios d'application : (1) l'utilisateur soumet des images, du texte ou des documents d'étiquettes pour contrôle de conformité ; (2) l'utilisateur mentionne des mots-clés tels que 'contrôle d'étiquetage', 'conformité d'étiquette', 'vérification de l'étiquetage nutritionnel', 'GB 7718' ou 'GB 28050' ; (3) l'utilisateur demande de vérifier si l'étiquette comporte des mentions manquantes, des erreurs ou des risques d'infraction. Prend en charge les versions 2011 et 2025 des normes." }, "ja": { "name": "food-label-compliance", "summary": "包装食品（プレパッケージ食品）ラベルのコンプライアンス審査 skill。GB 7718（包装食品ラベル通則）および GB 28050（包装食品栄養ラベル通則）への適合性を審査します。適用場面：(1) ユーザーが食品ラベルの画像・テキスト・文書を提出して適合審査を求める場合；(2)「食品ラベル審査」「ラベルコンプライアンス」「栄養ラベル審査」「GB 7718」「GB 28050」などのキーワードに言及する場合；(3) ラベルの記載漏れ・誤り・違反リスクの有無を確認したい場合。2011 年版と 2025 年版の両規格に対応します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:case-dashboard", "slug": "skill-case-dashboard", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Lawyer's case dashboard. A local zero-dependency dashboard service (case overview, three-state task board, deadline alerts, progress write-back) plus weekly analytical review. Use when the user says 'open the dashboard', 'dashboard', 'case panel', 'check case progress', or 'give me a weekly case ana" }, "human": { "zh-CN": { "name": "case-dashboard", "summary": "律师案件看板。本地零依赖看板服务（案件总览、任务三态看板、期限预警、进度写回）＋周研判分析。当用户说\"打开看板/dashboard/案件面板/看下案件进度/给我案件周报研判\"时使用。案件状态写入与上下文管理用 case-progress（/progress 命令也归它）。不要用于：案件实体分析、文书起草、法律研究。" }, "en": { "name": "case-dashboard", "summary": "Lawyer's case dashboard. A local zero-dependency dashboard service (case overview, three-state task board, deadline alerts, progress write-back) plus weekly analytical review. Use when the user says 'open the dashboard', 'dashboard', 'case panel', 'check case progress', or 'give me a weekly case analysis'. Case status writes and context management use case-progress (the /progress command also belongs to it). Not for: substantive case analysis, document drafting, or legal research." }, "es": { "name": "case-dashboard", "summary": "Panel de casos para abogados. Servicio de panel local sin dependencias (vista general de casos, tablero de tareas de tres estados, alertas de plazos, reescritura de progreso) más análisis semanal. Usar cuando el usuario dice 'abre el panel', 'dashboard', 'panel de casos', 'mira el avance del caso' o 'dame el análisis semanal de casos'. La escritura de estados del caso y la gestión de contexto usan case-progress (el comando /progress también le pertenece). No usar para: análisis sustantivo del caso, redacción de documentos ni investigación jurídica." }, "fr": { "name": "case-dashboard", "summary": "Tableau de bord des dossiers pour avocats. Service de tableau de bord local sans dépendance (vue d'ensemble des dossiers, tableau de tâches à trois états, alertes de délais, réécriture de l'avancement) plus analyse hebdomadaire. À utiliser lorsque l'utilisateur dit 'ouvre le tableau de bord', 'dashboard', 'panneau des dossiers', 'montre l'avancement' ou 'donne-moi l'analyse hebdomadaire'. L'écriture des statuts de dossier et la gestion du contexte passent par case-progress (la commande /progress lui appartient aussi). Ne pas utiliser pour : analyse substantive du dossier, rédaction d'actes, recherche juridique." }, "ja": { "name": "case-dashboard", "summary": "弁護士向け案件ダッシュボード。ローカルで依存なしの看板サービス（案件一覧、タスク三状態ボード、期限アラート、進捗書き戻し）＋週次研判分析。ユーザーが「看板を開いて」「dashboard」「案件パネル」「進捗を見て」「週次案件レポートの研判を」と言った場合に使用します。案件ステータスの書き込みとコンテキスト管理は case-progress を使用（/progress コマンドもこちらに帰属）。次の用途には使わない：案件実体分析、文書起草、リーガルリサーチ。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:case-progress", "slug": "skill-case-progress", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Case progress and status ledger management. The sole write engine for a case's case.yaml (adding tasks, advancing tasks, recording deadlines, updating case stage, validation, migration of legacy data), the progress load and write-back contract for Agent sessions, and reconciliation of case progress " }, "human": { "zh-CN": { "name": "case-progress", "summary": "案件进度与状态台账管理。案件 case.yaml 的唯一写入引擎（新增任务、推进任务、登记期限、更新案件阶段、校验、存量迁移），Agent 会话的进度加载与写回契约，以及案件进度监控对账。当用户或 Agent 需要\"更新案件进度/推进任务/登记期限/更新案件阶段/同步案件状态/校验或迁移案件数据/查看案件结构化状态\"时使用。不要用于：看板展示与周研判（用 case-dashboard）、案件建档初始化（用 new-case）。" }, "en": { "name": "case-progress", "summary": "Case progress and status ledger management. The sole write engine for a case's case.yaml (adding tasks, advancing tasks, recording deadlines, updating case stage, validation, migration of legacy data), the progress load and write-back contract for Agent sessions, and reconciliation of case progress monitoring. Use when the user or an Agent needs to 'update case progress', 'advance tasks', 'record deadlines', 'update case stage', 'sync case status', 'validate or migrate case data', or 'view structured case status'. Not for: dashboard display and weekly analysis (use case-dashboard), case file initialization (use new-case)." }, "es": { "name": "case-progress", "summary": "Gestión del libro de progreso y estado de casos. Motor único de escritura del case.yaml del caso (añadir tareas, avanzar tareas, registrar plazos, actualizar la etapa del caso, validación, migración de datos heredados), contrato de carga y reescritura de progreso para sesiones de Agent, y conciliación del monitoreo de progreso. Usar cuando el usuario o un Agent necesite 'actualizar el progreso del caso', 'avanzar tareas', 'registrar plazos', 'actualizar la etapa del caso', 'sincronizar el estado del caso', 'validar o migrar datos del caso' o 'ver el estado estructurado del caso'. No usar para: visualización en panel y análisis semanal (usar case-dashboard), inicialización del expediente (usar new-case)." }, "fr": { "name": "case-progress", "summary": "Gestion du registre d'avancement et d'état des dossiers. Moteur d'écriture unique du case.yaml du dossier (ajout de tâches, avancement des tâches, enregistrement des échéances, mise à jour de la phase, validation, migration des données héritées), contrat de chargement et de réécriture de l'avancement pour les sessions d'Agent, et rapprochement du suivi d'avancement. À utiliser lorsque l'utilisateur ou un Agent doit 'mettre à jour l'avancement du dossier', 'faire avancer des tâches', 'enregistrer des échéances', 'mettre à jour la phase', 'synchroniser le statut du dossier', 'valider ou migrer les données du dossier' ou 'consulter le statut structuré du dossier'. Ne pas utiliser pour : affichage tableau de bord et analyse hebdomadaire (utiliser case-dashboard), initialisation du dossier (utiliser new-case)." }, "ja": { "name": "case-progress", "summary": "案件の進捗・ステータス台帳管理。案件の case.yaml の唯一の書き込みエンジン（タスク追加、タスク推進、期限登録、案件フェーズ更新、検証、既存データ移行）、Agent セッションの進捗ロードと書き戻し契約、案件進捗モニタリングの照合を担います。ユーザーまたは Agent が「案件進捗の更新」「タスクの推進」「期限の登録」「案件フェーズの更新」「案件ステータスの同期」「案件データの検証・移行」「構造化ステータスの表示」を必要とする場合に使用します。次の用途には使わない：看板表示と週次研判（case-dashboard を使用）、案件ファイルの初期化（new-case を使用）。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:code2patent", "slug": "skill-code2patent", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Extracts technical implementation evidence from developed code projects, generates algorithm/software specification-style technical disclosure documents around candidate patent solutions, and, via a two-step method of 'claim layout card → invention patent first draft', further produces drafting mate" }, "human": { "zh-CN": { "name": "code2patent", "summary": "从已开发代码项目中提取技术实现证据，围绕候选专利方案生成算法/软件类说明书式技术交底书，并以“权利要求布局卡 → 发明专利初稿”两步法继续生成接近可申报版的中国发明专利起草材料。触发场景包括：读取代码仓库后撰写技术交底书、将人工总结的专利方案映射到具体实现、从代码中挖掘可专利技术方案、为专利代理师准备权利要求布局和发明专利初稿。" }, "en": { "name": "code2patent", "summary": "Extracts technical implementation evidence from developed code projects, generates algorithm/software specification-style technical disclosure documents around candidate patent solutions, and, via a two-step method of 'claim layout card → invention patent first draft', further produces drafting materials for Chinese invention patents close to a filable version. Trigger scenarios include: writing technical disclosures after reading a code repository, mapping manually summarized patent solutions to concrete implementations, mining patentable technical solutions from code, and preparing claim layouts and invention patent first drafts for patent agents." }, "es": { "name": "code2patent", "summary": "Extrae evidencia de implementación técnica de proyectos de código desarrollado, genera documentos de divulgación técnica estilo especificación para algoritmos/software en torno a soluciones de patente candidatas y, mediante el método de dos pasos 'tarjeta de disposición de reivindicaciones → borrador inicial de patente de invención', produce materiales de redacción de patentes de invención chinas cercanos a una versión presentable. Escenarios de activación: redactar divulgaciones técnicas tras leer un repositorio de código, mapear soluciones de patente resumidas manualmente a implementaciones concretas, extraer de código soluciones técnicas patentables y preparar disposiciones de reivindicaciones y borradores de patente para agentes de patentes." }, "fr": { "name": "code2patent", "summary": "Extrait des preuves de mise en œuvre technique de projets de code développés, génère des documents de divulgation technique de type spécification pour algorithmes/logiciels autour de solutions de brevet candidates et, via une méthode en deux étapes 'carte d'agencement des revendications → premier brouillon de brevet d'invention', produit des matériaux de rédaction de brevets d'invention chinois proches d'une version déposable. Scénarios de déclenchement : rédiger une divulgation technique après lecture d'un dépôt de code, faire correspondre des solutions de brevet résumées manuellement à des implémentations concrètes, extraire du code des solutions techniques brevetables, préparer l'agencement des revendications et des brouillons de brevet pour les mandataires en brevets." }, "ja": { "name": "code2patent", "summary": "開発済みコードプロジェクトから技術実装のエビデンスを抽出し、候補特許方案に基づいてアルゴリズム／ソフトウェア系の明細書スタイルの技術開示書（技術交底書）を生成します。さらに「クレーム配置カード → 発明特許初稿」の 2 段階方式で、出願可能版に近い中国発明特許の起草材料を生成します。発動シーン：コードリポジトリを読んだ後の技術開示書の作成、手作業でまとめた特許方案の具体的実装へのマッピング、コードからの特許可能な技術方案の発掘、特許代理人向けのクレーム配置と発明特許初稿の準備。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:contract-copilot", "slug": "skill-contract-copilot", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Contract drafting and review assistant. Based on layered analysis and a four-step process, it outputs actionable risk checklists, drafting skeletons, revision suggestions, recommended wording, and review opinions, and supports both annotation and tracked-changes document processing. It should also b" }, "human": { "zh-CN": { "name": "contract-copilot", "summary": "合同起草与审查助手。基于分层分析与四步流程，输出可执行的风险清单、起草骨架、修改建议、推荐措辞和审查意见书，支持批注与修订两种文档处理方式。用户通过飞书或其他 IM 对话发送合同文件并要求审查或起草时，也应使用本 skill，并优先沿原会话回传修订版和审查报告。" }, "en": { "name": "contract-copilot", "summary": "Contract drafting and review assistant. Based on layered analysis and a four-step process, it outputs actionable risk checklists, drafting skeletons, revision suggestions, recommended wording, and review opinions, and supports both annotation and tracked-changes document processing. It should also be used when the user sends a contract file via Feishu or another IM chat and asks for review or drafting, with the revised version and review report returned in the original conversation as a priority." }, "es": { "name": "contract-copilot", "summary": "Asistente de redacción y revisión de contratos. Basado en análisis por capas y un proceso de cuatro pasos, genera listas de riesgos accionables, esqueletos de redacción, sugerencias de modificación, formulaciones recomendadas y dictámenes de revisión, y admite el procesamiento de documentos tanto con anotaciones como con control de cambios. También debe usarse cuando el usuario envía un contrato por chat de Feishu u otra mensajería instantánea y solicita revisión o redacción, devolviendo preferentemente la versión revisada y el informe de revisión en la conversación original." }, "fr": { "name": "contract-copilot", "summary": "Assistant de rédaction et de revue de contrats. Fondé sur une analyse par couches et un processus en quatre étapes, il produit des listes de risques actionnables, des squelettes de rédaction, des suggestions de modification, des formulations recommandées et des avis de revue, et prend en charge deux modes de traitement des documents : annotations et suivi des modifications. Il doit également être utilisé lorsque l'utilisateur envoie un contrat via Feishu ou une autre messagerie instantanée et demande une revue ou une rédaction, en renvoyant en priorité la version révisée et le rapport de revue dans la conversation d'origine." }, "ja": { "name": "contract-copilot", "summary": "契約の起草・レビューアシスタント。階層分析と 4 ステップのプロセスに基づき、実行可能なリスクリスト、起草スケルトン、修正提案、推奨文言、レビュー意見書を出力し、注記（コメント）と変更履歴（リビジョン）の 2 つの文書処理方式に対応します。ユーザーが Feishu などの IM チャットで契約ファイルを送り、レビューや起草を依頼する場合にも本 skill を使用し、修正版とレビューレポートは元の会話で優先的に返送します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:legal-case-analysis", "slug": "skill-legal-case-analysis", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Use this skill when the user needs legal analysis, case assessment, risk evaluation, litigation strategy, criminal case assessment and review-for-prosecution analysis, non-litigation risk judgment, client explanations, or legal analysis reports based on case materials, consultation materials, contra" }, "human": { "zh-CN": { "name": "legal-case-analysis", "summary": "本技能应在用户需要基于案件材料、咨询材料、合同资料、证据材料或检索结果进行法律分析、案件研判、风险评估、诉讼策略、刑事案件研判与审查起诉分析、非诉风险判断、客户说明或法律分析报告时使用。它是法律任务的前置分析引擎，不要求每次都生成正式报告。不要用于：单纯 OCR、单纯法条案例检索、单纯 Word 排版转换、仅需包装服务方案且无需新增法律分析的文档生成。" }, "en": { "name": "legal-case-analysis", "summary": "Use this skill when the user needs legal analysis, case assessment, risk evaluation, litigation strategy, criminal case assessment and review-for-prosecution analysis, non-litigation risk judgment, client explanations, or legal analysis reports based on case materials, consultation materials, contract documents, evidence, or retrieval results. It is the upstream analysis engine for legal tasks and does not require generating a formal report every time. Not for: OCR alone, statute/case retrieval alone, Word layout conversion alone, or document generation that only packages a service proposal without new legal analysis." }, "es": { "name": "legal-case-analysis", "summary": "Usar esta skill cuando el usuario necesite análisis jurídico, valoración del caso, evaluación de riesgos, estrategia litigiosa, valoración penal y análisis de la revisión acusatoria, juicios de riesgo no contencioso, explicaciones al cliente o informes de análisis jurídico a partir de materiales del caso, de consulta, contractuales, de prueba o de resultados de búsqueda. Es el motor de análisis previo de las tareas jurídicas y no exige generar un informe formal cada vez. No usar para: OCR aislado, búsqueda de normas y casos aislada, conversión de maquetación Word aislada, o generación de documentos que solo empaquete una propuesta de servicios sin nuevo análisis jurídico." }, "fr": { "name": "legal-case-analysis", "summary": "Utiliser cette compétence lorsque l'utilisateur a besoin d'une analyse juridique, d'une appréciation du dossier, d'une évaluation des risques, d'une stratégie contentieuse, de l'appréciation d'affaires pénales et de l'analyse du réquisitoire, d'un jugement de risques non contentieux, d'explications au client ou de rapports d'analyse juridique à partir des pièces du dossier, documents de consultation, données contractuelles, preuves ou résultats de recherche. C'est le moteur d'analyse amont des tâches juridiques ; il n'exige pas un rapport formel à chaque fois. Ne pas utiliser pour : OCR seul, recherche de textes et jurisprudence seule, conversion de mise en page Word seule, ou génération de documents se bornant à mettre en forme une proposition de service sans nouvelle analyse juridique." }, "ja": { "name": "legal-case-analysis", "summary": "案件資料、相談資料、契約資料、証拠資料、または検索結果に基づく法律分析、案件研判、リスク評価、訴訟戦略、刑事事件の研判・起訴審査分析、非訴訟リスク判断、顧客向け説明、法律分析報告書が必要な場合に使用します。法律タスクの前段分析エンジンであり、毎回正式な報告書を生成する必要はありません。次の用途には使わない：単純な OCR、単純な条文・判例検索、単純な Word レイアウト変換、新しい法律分析を伴わないサービス提案の体裁整えだけの文書生成。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:legal-proposal-generator", "slug": "skill-legal-proposal-generator", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Generates various legal service documents from case materials or communication records (litigation proposals, consultation reports, non-litigation proposals, recommendation letters, communication reports, case-closing reports, case analysis summaries, etc.). Use this skill when the user needs to tur" }, "human": { "zh-CN": { "name": "legal-proposal-generator", "summary": "根据案件材料或沟通记录生成各类法律服务文档（诉讼方案、咨询报告、非诉方案、建议书、沟通报告、结案汇报、案件分析摘要等）。本技能应在用户需要将案件材料、咨询记录或沟通内容整理为专业法律文档时使用。" }, "en": { "name": "legal-proposal-generator", "summary": "Generates various legal service documents from case materials or communication records (litigation proposals, consultation reports, non-litigation proposals, recommendation letters, communication reports, case-closing reports, case analysis summaries, etc.). Use this skill when the user needs to turn case materials, consultation records, or communications into professional legal documents." }, "es": { "name": "legal-proposal-generator", "summary": "Genera diversos documentos de servicios jurídicos a partir de materiales del caso o registros de comunicación (propuestas litigiosas, informes de consulta, propuestas no contenciosas, cartas de recomendación, informes de comunicación, informes de cierre, resúmenes de análisis del caso, etc.). Usar esta skill cuando el usuario necesite convertir materiales del caso, registros de consulta o comunicaciones en documentos jurídicos profesionales." }, "fr": { "name": "legal-proposal-generator", "summary": "Génère divers documents de services juridiques à partir des pièces du dossier ou des échanges (propositions contentieuses, rapports de consultation, propositions non contentieuses, notes de recommandation, rapports d'échanges, rapports de clôture, synthèses d'analyse du dossier, etc.). Utiliser cette compétence lorsque l'utilisateur doit transformer des pièces du dossier, des enregistrements de consultation ou des échanges en documents juridiques professionnels." }, "ja": { "name": "legal-proposal-generator", "summary": "案件資料やコミュニケーション記録から各種法律サービス文書（訴訟方案、相談報告書、非訴訟方案、提案書、連絡報告書、結案報告、案件分析サマリーなど）を生成します。ユーザーが案件資料、相談記録、やり取りの内容をプロフェッショナルな法律文書に整理する必要がある場合に使用します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:legal-qa-extractor", "slug": "skill-legal-qa-extractor", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Extracts valuable legal Q&A pairs from lawyer-client communications to produce structured knowledge base content. Use this skill when the user needs to organize client consultation records, extract reusable legal knowledge from conversations, create Q&A knowledge bases, or prepare content marketing " }, "human": { "zh-CN": { "name": "legal-qa-extractor", "summary": "从律师与客户沟通记录中提取有价值的法律问答对，生成结构化知识库内容。本技能应在用户需要整理客户咨询记录、从对话中提取可复用法律知识、创建问答知识库、或准备内容营销素材时使用。支持严格客户信息脱敏处理。" }, "en": { "name": "legal-qa-extractor", "summary": "Extracts valuable legal Q&A pairs from lawyer-client communications to produce structured knowledge base content. Use this skill when the user needs to organize client consultation records, extract reusable legal knowledge from conversations, create Q&A knowledge bases, or prepare content marketing material. Supports strict client information de-identification." }, "es": { "name": "legal-qa-extractor", "summary": "Extrae pares valiosos de preguntas y respuestas jurídicas de las comunicaciones entre abogado y cliente para generar contenido estructurado de base de conocimiento. Usar esta skill cuando el usuario necesite organizar registros de consultas de clientes, extraer conocimiento jurídico reutilizable de conversaciones, crear bases de conocimiento de preguntas y respuestas, o preparar material de marketing de contenidos. Admite desidentificación estricta de la información del cliente." }, "fr": { "name": "legal-qa-extractor", "summary": "Extrait des paires questions-réponses juridiques à forte valeur des échanges entre avocat et client afin de produire du contenu structuré pour la base de connaissances. Utiliser cette compétence lorsque l'utilisateur doit organiser les enregistrements de consultation client, extraire des connaissances juridiques réutilisables de conversations, créer des bases de connaissances questions-réponses ou préparer des supports de marketing de contenu. Prend en charge l'anonymisation stricte des informations client." }, "ja": { "name": "legal-qa-extractor", "summary": "弁護士と顧客のコミュニケーション記録から価値の高い法律 Q&A ペアを抽出し、構造化されたナレッジベースコンテンツを生成します。ユーザーが顧客相談記録の整理、会話からの再利用可能な法律ナレッジの抽出、Q&A ナレッジベースの作成、コンテンツマーケティング素材の準備を必要とする場合に使用します。顧客情報の厳格な匿名化処理に対応します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:legal-text-format", "slug": "skill-legal-text-format", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Converts legal texts (statutes or cases) into well-formed Markdown and removes promotional redundant content. Use this skill when the user needs to process statutes (e.g., the Civil Code, the Criminal Law), organize cases (e.g., Supreme People's Court typical cases, judgment documents), or format le" }, "human": { "zh-CN": { "name": "legal-text-format", "summary": "将法律文本（法律条文或法律案例）转换为规范的 Markdown 格式，删除推广冗余信息。本技能应在用户需要处理法律条文（如民法典、刑法等）、整理法律案例（如最高法典型案例、裁判文书等）、或从粘贴文本中格式化法律文档时使用。注意：本技能只负责格式化和内容清理，不包含内容抓取能力。内容获取应由其他 skill（如 wechat-article-fetch）完成，AI 会自动判断技能协作顺序。内置 archive 机制，所有结果自包含存储在 skill 目录内。" }, "en": { "name": "legal-text-format", "summary": "Converts legal texts (statutes or cases) into well-formed Markdown and removes promotional redundant content. Use this skill when the user needs to process statutes (e.g., the Civil Code, the Criminal Law), organize cases (e.g., Supreme People's Court typical cases, judgment documents), or format legal documents from pasted text. Note: this skill only handles formatting and content cleanup; it has no content-scraping capability. Content fetching should be done by other skills (e.g., wechat-article-fetch), and the AI will automatically determine the collaboration order. Includes a built-in archive mechanism; all results are self-contained and stored within the skill directory." }, "es": { "name": "legal-text-format", "summary": "Convierte textos jurídicos (normas o casos) a formato Markdown normalizado y elimina el contenido promocional redundante. Usar esta skill cuando el usuario necesite procesar normas (p. ej., el Código Civil, el Código Penal), organizar casos (p. ej., casos típicos del Tribunal Popular Supremo, resoluciones judiciales) o formatear documentos jurídicos a partir de texto pegado. Nota: esta skill solo se encarga del formato y la limpieza de contenido; no incluye capacidad de captura de contenido. La obtención de contenido debe realizarla otra skill (p. ej., wechat-article-fetch), y la IA determinará automáticamente el orden de colaboración. Incluye un mecanismo de archivo integrado; todos los resultados se almacenan autocontenidos dentro del directorio de la skill." }, "fr": { "name": "legal-text-format", "summary": "Convertit les textes juridiques (textes de loi ou affaires) en Markdown normalisé et supprime le contenu promotionnel redondant. Utiliser cette compétence lorsque l'utilisateur doit traiter des textes de loi (p. ex. le Code civil, le droit pénal), organiser des affaires (p. ex. affaires types de la Cour populaire suprême, décisions de justice) ou mettre en forme des documents juridiques à partir de texte collé. Remarque : cette compétence ne gère que la mise en forme et le nettoyage du contenu ; elle n'inclut pas de capacité de récupération de contenu. L'obtention du contenu doit être faite par une autre compétence (p. ex. wechat-article-fetch), et l'IA déterminera automatiquement l'ordre de collaboration. Mécanisme d'archive intégré : tous les résultats sont autoportants et stockés dans le répertoire de la compétence." }, "ja": { "name": "legal-text-format", "summary": "法律テキスト（条文または判例）を整形された Markdown 形式に変換し、宣伝的な冗長情報を削除します。ユーザーが条文（例：民法典、刑法など）の処理、判例（例：最高人民法院の典型事例、裁判文書など）の整理、貼り付けテキストからの法律文書の整形を必要とする場合に使用します。注意：本 skill はフォーマットと内容のクリーンアップのみを担当し、コンテンツ取得機能は含みません。コンテンツの取得は他の skill（例：wechat-article-fetch）が行い、AI が協業順序を自動判断します。組み込みの archive 機構を備え、すべての結果は skill ディレクトリ内に自己完結で保存されます。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:legal-visualization", "slug": "skill-legal-visualization", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Legal Visualization. A skill for legal diagrams and chart generation in legal business scenarios; use when the user asks to organize case materials, contract materials, compliance matters, transaction arrangements, evidence chains, litigation processes, timelines, legal relationships, client reports" }, "human": { "zh-CN": { "name": "legal-visualization", "summary": "Legal Visualization。面向法律业务场景的法律图解与图表生成技能；当用户要求把案件材料、合同材料、合规事项、交易安排、证据链、诉讼流程、时间轴、法律关系、客户汇报、法律服务方案或律师团队工作整理成关系图、流程图、时间轴、证据链、风险图、路线图、PNG/SVG/PDF/.drawio 时使用；也兼容“法律可视化”“案件事实图”“法律关系图”等说法。先按受众、任务动词和路由规则筛选场景，再生成可交付图片，并保留 draw.io 源文件作为可编辑底稿。本技能不用于事实核验，也不替代法律结论判断。" }, "en": { "name": "legal-visualization", "summary": "Legal Visualization. A skill for legal diagrams and chart generation in legal business scenarios; use when the user asks to organize case materials, contract materials, compliance matters, transaction arrangements, evidence chains, litigation processes, timelines, legal relationships, client reports, legal service proposals, or lawyer team work into relationship diagrams, flowcharts, timelines, evidence chains, risk maps, roadmaps, or PNG/SVG/PDF/.drawio outputs; also compatible with phrasings like 'legal visualization', 'case fact diagram', or 'legal relationship diagram'. Scenarios are first screened by audience, task verbs, and routing rules, then deliverable images are generated, with draw.io source files retained as editable drafts. This skill is not for fact verification and does not substitute for legal conclusion judgment." }, "es": { "name": "legal-visualization", "summary": "Legal Visualization. Skill de generación de diagramas y gráficos jurídicos para escenarios de negocio legal; usar cuando el usuario pida organizar materiales del caso, materiales contractuales, asuntos de cumplimiento, arreglos transaccionales, cadenas probatorias, procesos litigiosos, líneas de tiempo, relaciones jurídicas, informes al cliente, propuestas de servicios jurídicos o trabajo del equipo de abogados en diagramas de relaciones, diagramas de flujo, líneas de tiempo, cadenas probatorias, mapas de riesgo, hojas de ruta o salidas PNG/SVG/PDF/.drawio; también acepta formulaciones como 'visualización jurídica', 'diagrama de hechos del caso' o 'diagrama de relaciones jurídicas'. Primero se filtran los escenarios por audiencia, verbos de tarea y reglas de enrutamiento; luego se generan las imágenes entregables, conservando los archivos fuente draw.io como borradores editables. Esta skill no sirve para verificar hechos ni sustituye el juicio de conclusiones jurídicas." }, "fr": { "name": "legal-visualization", "summary": "Legal Visualization. Compétence de génération de schémas et graphiques juridiques pour les scénarios métier juridiques ; à utiliser lorsque l'utilisateur demande d'organiser des pièces de dossier, des documents contractuels, des sujets de conformité, des montages transactionnels, des chaînes de preuves, des processus contentieux, des frises chronologiques, des relations juridiques, des rapports client, des propositions de services juridiques ou le travail de l'équipe d'avocats en diagrammes de relations, organigrammes, frises, chaînes de preuves, cartes de risques, feuilles de route ou sorties PNG/SVG/PDF/.drawio ; accepte aussi les formulations 'visualisation juridique', 'schéma des faits de l'affaire' ou 'schéma des relations juridiques'. Les scénarios sont d'abord filtrés par audience, verbes de tâche et règles de routage, puis les images livrables sont générées, avec conservation des fichiers sources draw.io comme brouillons modifiables. Cette compétence ne sert pas à vérifier les faits et ne remplace pas l'appréciation des conclusions juridiques." }, "ja": { "name": "legal-visualization", "summary": "Legal Visualization。法律業務シーン向けの法律図解・図表生成 skill。ユーザーが案件資料、契約資料、コンプライアンス事項、取引設計、証拠チェーン、訴訟フロー、タイムライン、法律関係、顧客報告、法律サービス提案、弁護士チームの業務を、関係図・フローチャート・タイムライン・証拠チェーン・リスク図・ロードマップ・PNG/SVG/PDF/.drawio に整理するよう求めた場合に使用します。「法律の可視化」「事件事実図」「法律関係図」などの言い回しにも対応します。まずオーディエンス、タスク動詞、ルーティング規則でシーンを絞り込み、納品可能な画像を生成し、draw.io ソースファイルを編集可能な下書きとして保持します。本 skill は事実検証には使用せず、法律上の結論判断の代用にもなりません。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:litigation-analysis", "slug": "skill-litigation-analysis", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Litigation analysis tool - deep analysis of judgment documents, producing appeal/retrial decision support." }, "human": { "zh-CN": { "name": "litigation-analysis", "summary": "诉讼分析工具 - 判决书深度分析，生成上诉/再审决策支持" }, "en": { "name": "litigation-analysis", "summary": "Litigation analysis tool - deep analysis of judgment documents, producing appeal/retrial decision support." }, "es": { "name": "litigation-analysis", "summary": "Herramienta de análisis de litigios - análisis profundo de resoluciones judiciales, genera soporte de decisión para apelación / revisión." }, "fr": { "name": "litigation-analysis", "summary": "Outil d'analyse contentieuse - analyse approfondie des décisions de justice, produit une aide à la décision pour l'appel / le réexamen." }, "ja": { "name": "litigation-analysis", "summary": "訴訟分析ツール - 判決書の深度分析を行い、控訴・再審の意思決定サポートを生成します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:new-case", "slug": "skill-new-case", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Organizes case/consultation/trademark/patent materials into a standardized directory structure. Four presets: litigation cases (12 directories), prospective projects/consultations (2 directories), trademark cases (2 templates by business sub-type: registration, 7 directories including an independent" }, "human": { "zh-CN": { "name": "new-case", "summary": "将案件/咨询/商标/专利材料整理成标准化目录结构。支持诉讼案件（12目录）、潜在项目/咨询（2目录）、商标案件（按业务子类型选 2 个模板：注册 7 目录含独立商标注册证/异议复审无效 8 目录含独立商标注册证）、专利案件（按业务子类型选 3 个模板：申请 7 目录/复审无效 8 目录/检索其他 4 目录，目录顺序按项目进展）四种预设。本技能应在用户需要创建新案件、初始化案件目录结构、整理咨询材料、接收商标/专利新业务、或通过参数和自然语言指定案件编号、委托人、案件类型等信息快速创建案件时使用。不要用于：单独生成法律文书、进行法律研究、证据分析等非案件初始化任务。" }, "en": { "name": "new-case", "summary": "Organizes case/consultation/trademark/patent materials into a standardized directory structure. Four presets: litigation cases (12 directories), prospective projects/consultations (2 directories), trademark cases (2 templates by business sub-type: registration, 7 directories including an independent trademark registration certificate; opposition/review/invalidation, 8 directories including an independent trademark registration certificate), and patent cases (3 templates by business sub-type: application, 7 directories; review/invalidation, 8 directories; search and others, 4 directories, with directory order following project progress). Use this skill when the user needs to create a new case, initialize a case directory structure, organize consultation materials, onboard new trademark/patent business, or quickly create a case by specifying case number, client, case type, etc. via parameters and natural language. Not for: standalone legal document generation, legal research, evidence analysis, or other non-initialization tasks." }, "es": { "name": "new-case", "summary": "Organiza materiales de casos, consultas, marcas y patentes en una estructura de directorios estandarizada. Cuatro preajustes: casos litigiosos (12 directorios), proyectos potenciales/consultas (2 directorios), casos de marca (2 plantillas según subtipo de negocio: registro, 7 directorios con certificado de registro de marca independiente; oposición/revisión/nulidad, 8 directorios con certificado de registro de marca independiente) y casos de patente (3 plantillas según subtipo: solicitud, 7 directorios; revisión/nulidad, 8 directorios; búsqueda y otros, 4 directorios, con el orden de directorios según el avance del proyecto). Usar esta skill cuando el usuario necesite crear un caso nuevo, inicializar la estructura de directorios, organizar materiales de consulta, incorporar nuevos negocios de marca/patente, o crear rápidamente un caso indicando número de caso, cliente, tipo de caso, etc. mediante parámetros y lenguaje natural. No usar para: generación independiente de documentos jurídicos, investigación jurídica, análisis de pruebas ni otras tareas ajenas a la inicialización del caso." }, "fr": { "name": "new-case", "summary": "Organise les matériaux de dossiers, consultations, marques et brevets en une structure de répertoires normalisée. Quatre préréglages : dossiers contentieux (12 répertoires), projets potentiels/consultations (2 répertoires), dossiers de marque (2 modèles selon le sous-type d'affaires : enregistrement, 7 répertoires avec certificat de marque indépendant ; opposition/réexamen/nullité, 8 répertoires avec certificat de marque indépendant), dossiers de brevet (3 modèles selon le sous-type : demande, 7 répertoires ; réexamen/nullité, 8 répertoires ; recherche et autres, 4 répertoires, l'ordre des répertoires suivant l'avancement du projet). Utiliser cette compétence lorsque l'utilisateur doit créer un nouveau dossier, initialiser la structure de répertoires, organiser des matériaux de consultation, prendre en charge de nouvelles affaires de marque/brevet, ou créer rapidement un dossier en spécifiant numéro de dossier, client, type de dossier, etc. via des paramètres et du langage naturel. Ne pas utiliser pour : génération autonome de documents juridiques, recherche juridique, analyse de preuves ou autres tâches hors initialisation de dossier." }, "ja": { "name": "new-case", "summary": "案件・相談・商標・特許の資料を標準化されたディレクトリ構造に整理します。4 つのプリセット：訴訟案件（12 ディレクトリ）、見込み案件・相談（2 ディレクトリ）、商標案件（業務サブタイプで 2 テンプレート：登録 7 ディレクトリ（独立した商標登録証を含む）／異議・審判・無効 8 ディレクトリ（独立した商標登録証を含む））、特許案件（業務サブタイプで 3 テンプレート：出願 7 ディレクトリ／審判・無効 8 ディレクトリ／検索・その他 4 ディレクトリ、ディレクトリ順は案件の進展に従う）。ユーザーが新規案件の作成、案件ディレクトリ構造の初期化、相談資料の整理、商標・特許の新規受任、または案件番号・依頼者・案件タイプなどをパラメータと自然言語で指定して迅速に案件を作成する必要がある場合に使用します。次の用途には使わない：法律文書の単独生成、リーガルリサーチ、証拠分析など案件初期化以外のタスク。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:opc-legal-counsel", "slug": "skill-opc-legal-counsel", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "A legal triage and judgment skill for OPC (One Person Company - single-person company / solo founder + AI collaboration) and small/micro business operation scenarios. Use when the user consults on entity choice, separation of company and personal affairs, co-founders, financing and incentives, contr" }, "human": { "zh-CN": { "name": "opc-legal-counsel", "summary": "面向 OPC（One Person Company，一人公司 / 单人创业者 + AI 协同）和小微企业经营场景的法律业务分诊与判断技能。用户咨询主体选择、公私分离、联合创始人、融资与激励、合同履约、财税风险、AI 产品、数据与知识产权、用工、监管或争议时使用。负责识别主矛盾、联动风险、行动优先级和升级边界；凡涉及现行法条、效力状态、期限、税率、备案名录、地方政策、办理口径或平台规则，生成结构化检索任务并交由可用法律数据库、MCP 或官方来源核验，不以内置知识冒充最新规则。不要用于替代正式法律意见、深度合同批注、诉讼文书、商标申请、专利分析或纯法规检索。" }, "en": { "name": "opc-legal-counsel", "summary": "A legal triage and judgment skill for OPC (One Person Company - single-person company / solo founder + AI collaboration) and small/micro business operation scenarios. Use when the user consults on entity choice, separation of company and personal affairs, co-founders, financing and incentives, contract performance, fiscal/tax risks, AI products, data and IP, employment, regulation, or disputes. It identifies the primary conflict, linked risks, action priorities, and escalation boundaries; anything involving current statutes, validity status, deadlines, tax rates, filing lists, local policies, processing practices, or platform rules generates structured retrieval tasks delegated to available legal databases, MCP, or official sources - built-in knowledge is never passed off as the latest rules. Not for: replacing formal legal opinions, deep contract annotation, litigation documents, trademark applications, patent analysis, or pure statute retrieval." }, "es": { "name": "opc-legal-counsel", "summary": "Skill de triaje y juicio jurídico para escenarios de OPC (One Person Company - empresa unipersonal / emprendedor individual + colaboración con IA) y operación de micro y pequeñas empresas. Usar cuando el usuario consulte sobre elección de la forma jurídica, separación entre lo empresarial y lo personal, cofundadores, financiación e incentivos, cumplimiento contractual, riesgos fiscales, productos de IA, datos y propiedad intelectual, empleo, regulación o disputas. Se encarga de identificar la contradicción principal, los riesgos encadenados, las prioridades de acción y los límites de escalado; todo lo que implique normas vigentes, estado de vigencia, plazos, tasas impositivas, listados de registro, políticas locales, criterios de tramitación o reglas de plataforma genera tareas de búsqueda estructuradas que se delegan en bases de datos jurídicas disponibles, MCP o fuentes oficiales - el conocimiento integrado nunca se hace pasar por las reglas más recientes. No usar para: sustituir dictámenes jurídicos formales, anotación profunda de contratos, documentos procesales, solicitudes de marca, análisis de patentes o búsqueda normativa pura." }, "fr": { "name": "opc-legal-counsel", "summary": "Compétence de triage et d'appréciation juridique pour les scénarios OPC (One Person Company - société unipersonnelle / entrepreneur solo + collaboration IA) et l'exploitation de très petites entreprises. À utiliser lorsque l'utilisateur consulte sur le choix de la structure, la séparation des sphères professionnelle et privée, les cofondateurs, le financement et les incitations, l'exécution des contrats, les risques fiscaux, les produits IA, les données et la propriété intellectuelle, l'emploi, la régulation ou les litiges. Elle identifie la contradiction principale, les risques liés, les priorités d'action et les limites d'escalade ; tout ce qui touche aux textes en vigueur, à l'état de validité, aux échéances, aux taux d'imposition, aux répertoires de déclaration, aux politiques locales, aux pratiques de traitement ou aux règles de plateforme génère des tâches de recherche structurées confiées aux bases de données juridiques disponibles, à MCP ou à des sources officielles - les connaissances intégrées ne se substituent jamais aux règles les plus récentes. Ne pas utiliser pour : remplacer des avis juridiques formels, l'annotation approfondie de contrats, les actes de procédure, les dépôts de marque, l'analyse de brevets ou la pure recherche réglementaire." }, "ja": { "name": "opc-legal-counsel", "summary": "OPC（One Person Company：一人会社／個人起業家＋AI 協働）と小規模事業者の経営シーン向けの法務トリアージ・判断 skill。主体選択、公私分離、共同創業者、資金調達とインセンティブ、契約履行、財務・税務リスク、AI 製品、データと知的財産、雇用、規制、紛争に関する相談時に使用します。主たる矛盾、連動リスク、行動の優先順位、エスカレーション境界の特定を担います。現行法令、効力状態、期限、税率、届出名簿、地方政策、処理基準、プラットフォーム規則に関わる事項は、構造化された検索タスクを生成して利用可能な法律データベース、MCP、または公式ソースによる検証に委ね、内蔵知識を最新規則の代わりとしません。次の用途には使わない：正式な法律意見の代替、契約の詳細注記、訴訟文書、商標出願、特許分析、純粋な法令検索。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:patent-analysis", "slug": "skill-patent-analysis", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "A structured preliminary analysis tool for Chinese invention and utility model patents. Use this skill when the user needs to break down or interpret claims, compare patent protection scopes, perform product infringement comparisons, equivalence analysis, stability and invalidation risk analysis, FT" }, "human": { "zh-CN": { "name": "patent-analysis", "summary": "面向中国发明和实用新型专利的结构化初步分析工具。本技能应在用户需要拆解或解释权利要求、比较专利保护范围、进行产品侵权比对、等同分析、稳定性与无效风险分析、FTO、规避设计、专利价值评估或制作专利分析图表时使用。不要用于：代替专利律师或专利代理师出具正式意见、仅下载/OCR专利、撰写专利申请文件，或在未确认法域与有效权利要求时给出确定性结论；外观设计近似判断需另行采用专门方法并人工复核。" }, "en": { "name": "patent-analysis", "summary": "A structured preliminary analysis tool for Chinese invention and utility model patents. Use this skill when the user needs to break down or interpret claims, compare patent protection scopes, perform product infringement comparisons, equivalence analysis, stability and invalidation risk analysis, FTO, design-arounds, patent value assessment, or produce patent analysis charts. Not for: replacing formal opinions of patent lawyers or patent agents, merely downloading/OCR-ing patents, drafting patent application documents, or giving definitive conclusions without confirming the jurisdiction and valid claims; design-similarity judgment for design patents requires a dedicated method and manual re-check." }, "es": { "name": "patent-analysis", "summary": "Herramienta de análisis preliminar estructurado para patentes de invención y modelos de utilidad chinos. Usar esta skill cuando el usuario necesite desglosar o interpretar reivindicaciones, comparar alcances de protección, realizar comparaciones de infracción de producto, análisis de equivalencia, análisis de riesgo de estabilidad e invalidación, FTO, diseños evasivos, valoración de patentes o gráficos de análisis de patentes. No usar para: sustituir el dictamen formal de abogados o agentes de patentes, solo descargar/hacer OCR de patentes, redactar documentos de solicitud de patente, o emitir conclusiones definitivas sin confirmar la jurisdicción y las reivindicaciones vigentes; el juicio de similitud de diseños industriales requiere un método específico y revisión manual." }, "fr": { "name": "patent-analysis", "summary": "Outil d'analyse préliminaire structurée pour les brevets d'invention et modèles d'utilité chinois. Utiliser cette compétence lorsque l'utilisateur doit décomposer ou interpréter des revendications, comparer des étendues de protection, mener des comparaisons de contrefaçon de produit, une analyse d'équivalence, une analyse des risques de stabilité et de nullité, du FTO, des contournements (design-arounds), une valorisation de brevets ou des graphiques d'analyse de brevets. Ne pas utiliser pour : remplacer l'avis formel d'un conseil ou mandataire en brevets, seulement télécharger/faire de l'OCR sur des brevets, rédiger des documents de dépôt, ou conclure de façon définitive sans avoir confirmé la juridiction et les revendications en vigueur ; l'appréciation de similitude des dessins et modèles exige une méthode dédiée et une relecture humaine." }, "ja": { "name": "patent-analysis", "summary": "中国の発明特許・実用新案特許向けの構造化された予備分析ツール。ユーザーがクレームの分解や解釈、特許保護範囲の比較、製品侵害比較、均等分析、安定性・無効リスク分析、FTO、回避設計、特許価値評価、特許分析図表の作成を必要とする場合に使用します。次の用途には使わない：特許弁護士や弁理士の正式意見の代替、特許のダウンロード/OCR のみ、特許出願文書の作成、法域と有効なクレームを確認しないままの確定的結論。意匠の類似判断は専用手法と人手での再確認が必要です。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:trademark-assistant", "slug": "skill-trademark-assistant", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "A skill for class planning, registrability pre-screening, and application material preparation for Chinese trademark applications. Based on the Nice Classification (international classification of trademarks) and citing Chinese laws and regulations. Use when receiving trademark consultations, when s" }, "human": { "zh-CN": { "name": "trademark-assistant", "summary": "面向中国商标申请的类别规划、可注册性初筛及申请材料准备技能。基于尼斯分类（国际商标分类），引用中国法律法规。本技能应在接到商标咨询、需要结构化输出建议与风险分级、或需要准备申请材料时使用。不要用于：替代正式法律意见、承诺注册成功率、处理复杂商标争议案件。" }, "en": { "name": "trademark-assistant", "summary": "A skill for class planning, registrability pre-screening, and application material preparation for Chinese trademark applications. Based on the Nice Classification (international classification of trademarks) and citing Chinese laws and regulations. Use when receiving trademark consultations, when structured advice with risk grading is needed, or when preparing application materials. Not for: replacing formal legal opinions, promising registration success rates, or handling complex trademark disputes." }, "es": { "name": "trademark-assistant", "summary": "Skill de planificación de clases, preselección de registrabilidad y preparación de materiales de solicitud para solicitudes de marca en China. Basado en la Clasificación de Niza (clasificación internacional de marcas) y cita la legislación china. Usar al recibir consultas de marca, cuando se necesiten consejos estructurados con gradación de riesgos, o al preparar materiales de solicitud. No usar para: sustituir dictámenes jurídicos formales, prometer tasas de éxito de registro ni gestionar disputas complejas de marca." }, "fr": { "name": "trademark-assistant", "summary": "Compétence de planification des classes, de présélection de la registrabilité et de préparation des dossiers de demande pour les dépôts de marques en Chine. Fondée sur la Classification de Nice (classification internationale des marques) et citant la législation chinoise. À utiliser lors de consultations sur les marques, en cas de besoin de conseils structurés avec gradation des risques, ou pour préparer les dossiers de demande. Ne pas utiliser pour : remplacer un avis juridique formel, promettre un taux de succès d'enregistrement, ou traiter des litiges complexes de marque." }, "ja": { "name": "trademark-assistant", "summary": "中国の商標出願向けの区分プランニング、登録可能性の予備スクリーニング、出願資料準備の skill。ニース分類（国際商標分類）に基づき、中国の法令を引用します。商標相談の受付、構造化された助言とリスク等級付けが必要な場合、出願資料の準備が必要な場合に使用します。次の用途には使わない：正式な法律意見の代替、登録成功率の保証、複雑な商標係争案件の処理。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:wechat-article-to-markdown", "slug": "skill-wechat-article-to-markdown", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Converts legal texts (statutes or cases) into well-formed Markdown and removes promotional redundant content. Use this skill when the user needs to process statutes (e.g., the Civil Code, the Criminal Law), organize cases (e.g., Supreme People's Court typical cases, judgment documents), or format le" }, "human": { "zh-CN": { "name": "wechat-article-to-markdown", "summary": "将法律文本（法律条文或法律案例）转换为规范的 Markdown 格式，删除推广冗余信息。本技能应在用户需要处理法律条文（如民法典、刑法等）、整理法律案例（如最高法典型案例、裁判文书等）、或从粘贴文本中格式化法律文档时使用。注意：本技能只负责格式化和内容清理，不包含内容抓取能力。内容获取应由其他 skill（如 wechat-article-fetch）完成，AI 会自动判断技能协作顺序。" }, "en": { "name": "wechat-article-to-markdown", "summary": "Converts legal texts (statutes or cases) into well-formed Markdown and removes promotional redundant content. Use this skill when the user needs to process statutes (e.g., the Civil Code, the Criminal Law), organize cases (e.g., Supreme People's Court typical cases, judgment documents), or format legal documents from pasted text. Note: this skill only handles formatting and content cleanup; it has no content-scraping capability. Content fetching should be done by other skills (e.g., wechat-article-fetch), and the AI will automatically determine the collaboration order. Includes a built-in archive mechanism; all results are self-contained and stored within the skill directory." }, "es": { "name": "wechat-article-to-markdown", "summary": "Convierte textos jurídicos (normas o casos) a formato Markdown normalizado y elimina el contenido promocional redundante. Usar esta skill cuando el usuario necesite procesar normas (p. ej., el Código Civil, el Código Penal), organizar casos (p. ej., casos típicos del Tribunal Popular Supremo, resoluciones judiciales) o formatear documentos jurídicos a partir de texto pegado. Nota: esta skill solo se encarga del formato y la limpieza de contenido; no incluye capacidad de captura de contenido. La obtención de contenido debe realizarla otra skill (p. ej., wechat-article-fetch), y la IA determinará automáticamente el orden de colaboración. Incluye un mecanismo de archivo integrado; todos los resultados se almacenan autocontenidos dentro del directorio de la skill." }, "fr": { "name": "wechat-article-to-markdown", "summary": "Convertit les textes juridiques (textes de loi ou affaires) en Markdown normalisé et supprime le contenu promotionnel redondant. Utiliser cette compétence lorsque l'utilisateur doit traiter des textes de loi (p. ex. le Code civil, le droit pénal), organiser des affaires (p. ex. affaires types de la Cour populaire suprême, décisions de justice) ou mettre en forme des documents juridiques à partir de texte collé. Remarque : cette compétence ne gère que la mise en forme et le nettoyage du contenu ; elle n'inclut pas de capacité de récupération de contenu. L'obtention du contenu doit être faite par une autre compétence (p. ex. wechat-article-fetch), et l'IA déterminera automatiquement l'ordre de collaboration. Mécanisme d'archive intégré : tous les résultats sont autoportants et stockés dans le répertoire de la compétence." }, "ja": { "name": "wechat-article-to-markdown", "summary": "法律テキスト（条文または判例）を整形された Markdown 形式に変換し、宣伝的な冗長情報を削除します。ユーザーが条文（例：民法典、刑法など）の処理、判例（例：最高人民法院の典型事例、裁判文書など）の整理、貼り付けテキストからの法律文書の整形を必要とする場合に使用します。注意：本 skill はフォーマットと内容のクリーンアップのみを担当し、コンテンツ取得機能は含みません。コンテンツの取得は他の skill（例：wechat-article-fetch）が行い、AI が協業順序を自動判断します。組み込みの archive 機構を備え、すべての結果は skill ディレクトリ内に自己完結で保存されます。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:case-materials-to-legal-docs", "slug": "skill-case-materials-to-legal-docs", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Generates various legal service documents from case materials or communication records (litigation proposals, consultation reports, non-litigation proposals, recommendation letters, communication reports, etc.). Use this skill when the user needs to turn case materials, consultation records, or comm" }, "human": { "zh-CN": { "name": "case-materials-to-legal-docs", "summary": "根据案件材料或沟通记录生成各类法律服务文档（诉讼方案、咨询报告、非诉方案、建议书、沟通报告等）。本技能应在用户需要将案件材料、咨询记录或沟通内容整理为专业法律文档时使用。" }, "en": { "name": "case-materials-to-legal-docs", "summary": "Generates various legal service documents from case materials or communication records (litigation proposals, consultation reports, non-litigation proposals, recommendation letters, communication reports, etc.). Use this skill when the user needs to turn case materials, consultation records, or communications into professional legal documents." }, "es": { "name": "case-materials-to-legal-docs", "summary": "Genera diversos documentos de servicios jurídicos a partir de materiales del caso o registros de comunicación (propuestas litigiosas, informes de consulta, propuestas no contenciosas, cartas de recomendación, informes de comunicación, etc.). Usar esta skill cuando el usuario necesite convertir materiales del caso, registros de consulta o comunicaciones en documentos jurídicos profesionales." }, "fr": { "name": "case-materials-to-legal-docs", "summary": "Génère divers documents de services juridiques à partir des pièces du dossier ou des échanges (propositions contentieuses, rapports de consultation, propositions non contentieuses, notes de recommandation, rapports d'échanges, etc.). Utiliser cette compétence lorsque l'utilisateur doit transformer des pièces du dossier, des enregistrements de consultation ou des échanges en documents juridiques professionnels." }, "ja": { "name": "case-materials-to-legal-docs", "summary": "案件資料やコミュニケーション記録から各種法律サービス文書（訴訟方案、相談報告書、非訴訟方案、提案書、連絡報告書など）を生成します。ユーザーが案件資料、相談記録、やり取りの内容をプロフェッショナルな法律文書に整理する必要がある場合に使用します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:litigation-analysis-tool", "slug": "skill-litigation-analysis-tool", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Litigation analysis tool - supports folder batch processing and deep analysis of judgment documents, producing appeal/retrial decision support." }, "human": { "zh-CN": { "name": "litigation-analysis-tool", "summary": "诉讼分析工具 - 支持文件夹批量处理，判决书深度分析，生成上诉/再审决策支持" }, "en": { "name": "litigation-analysis-tool", "summary": "Litigation analysis tool - supports folder batch processing and deep analysis of judgment documents, producing appeal/retrial decision support." }, "es": { "name": "litigation-analysis-tool", "summary": "Herramienta de análisis de litigios - admite procesamiento por lotes de carpetas, análisis profundo de resoluciones judiciales y genera soporte de decisión para apelación / revisión." }, "fr": { "name": "litigation-analysis-tool", "summary": "Outil d'analyse contentieuse - prend en charge le traitement par lot de dossiers, l'analyse approfondie des décisions de justice et produit une aide à la décision pour l'appel / le réexamen." }, "ja": { "name": "litigation-analysis-tool", "summary": "訴訟分析ツール - フォルダ単位の一括処理に対応し、判決書を深度分析して、控訴・再審の意思決定サポートを生成します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:court-sms", "slug": "skill-court-sms", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Use this skill when the user receives court SMS messages (document service, filing notifications, hearing reminders, etc.); it automatically extracts case numbers, parties, and download links, downloads the documents, and archives them into the corresponding case directory. It writes sensitive litig" }, "human": { "zh-CN": { "name": "court-sms", "summary": "本技能应在用户收到法院短信（文书送达、立案通知、开庭提醒等）时使用，自动提取案号、当事人、下载链接，下载文书并归档到对应案件目录。会落盘敏感诉讼数据到本地，请确认目标位置私密。" }, "en": { "name": "court-sms", "summary": "Use this skill when the user receives court SMS messages (document service, filing notifications, hearing reminders, etc.); it automatically extracts case numbers, parties, and download links, downloads the documents, and archives them into the corresponding case directory. It writes sensitive litigation data to local disk, so confirm the target location is private." }, "es": { "name": "court-sms", "summary": "Usar esta skill cuando el usuario reciba SMS del tribunal (notificación de documentos, avisos de presentación de demanda, recordatorios de audiencia, etc.); extrae automáticamente números de caso, partes y enlaces de descarga, descarga los documentos y los archiva en el directorio del caso correspondiente. Escribe datos sensibles de litigio en el disco local, así que confirme que la ubicación de destino es privada." }, "fr": { "name": "court-sms", "summary": "Utiliser cette compétence lorsque l'utilisateur reçoit des SMS du tribunal (signification de pièces, notifications de saisine, rappels d'audience, etc.) ; elle extrait automatiquement les numéros de dossier, les parties et les liens de téléchargement, télécharge les pièces et les archive dans le répertoire du dossier correspondant. Elle écrit des données de contentieux sensibles sur le disque local : vérifiez que l'emplacement cible est privé." }, "ja": { "name": "court-sms", "summary": "ユーザーが裁判所からの SMS（文書送達、立案通知、開廷リマインダーなど）を受信した際に使用します。事件番号、当事者、ダウンロードリンクを自動抽出し、文書をダウンロードして対応する案件ディレクトリにアーカイブします。機微性の高い訴訟データをローカルディスクに書き込むため、保存先がプライベートであることを確認してください。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:tencent-ima-copilot-legal-consultation", "slug": "skill-tencent-ima-copilot-legal-consultation", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "legalskill" ], "canonical": { "brief": "Provides preliminary diagnosis and action strategies for legal questions in the staged case-handling mindset of a senior lawyer, combined with RAG retrieval from a legal knowledge base. Triggered when the user describes a legal dispute, asks 'what to do', 'can I sue', 'will I go to jail', asks about" }, "human": { "zh-CN": { "name": "tencent-ima-copilot-legal-consultation", "summary": "以资深律师分阶段办案思维，结合 RAG 法律知识库检索，为用户提供法律问题初步诊断与行动策略。当用户描述法律纠纷、咨询\"怎么办\"\"能不能告\"\"会不会坐牢\"、询问欠钱不还/被起诉/打官司/维权/赔偿/工伤/交通事故/租房纠纷/遗产继承/离婚/合同纠纷/消费者权益/公司劳动争议等场景，或要求分析合同/侵权/劳动/家事/行政/公司股权问题，或上传法律文件时触发。不适用于出具正式法律意见书、代写文书或替代律师出庭。支持分层精读策略、多轮换词搜索、结构化引用清单（含效力状态标注）、快速/专家双模式。内置反幻觉保障机制（绝对溯源+时效性验证+幻觉自检清单），确保法条引用可溯源、案例裁判经核验、失效法条自动标注。" }, "en": { "name": "tencent-ima-copilot-legal-consultation", "summary": "Provides preliminary diagnosis and action strategies for legal questions in the staged case-handling mindset of a senior lawyer, combined with RAG retrieval from a legal knowledge base. Triggered when the user describes a legal dispute, asks 'what to do', 'can I sue', 'will I go to jail', asks about unpaid debts / being sued / going to court / rights protection / compensation / work injury / traffic accidents / rental disputes / inheritance / divorce / contract disputes / consumer rights / company labor disputes, requests analysis of contract / tort / labor / family / administrative / company equity issues, or uploads legal documents. Not suitable for issuing formal legal opinions, ghostwriting documents, or substituting for a lawyer in court. Supports a layered close-reading strategy, multi-round synonym search, structured citation lists (with validity status annotations), and quick/expert dual modes. Built-in anti-hallucination safeguards (absolute source tracing + timeliness verification + hallucination self-check checklist) ensure statute citations are traceable, case rulings are verified, and repealed provisions are automatically flagged." }, "es": { "name": "tencent-ima-copilot-legal-consultation", "summary": "Ofrece diagnóstico preliminar y estrategias de actuación para problemas jurídicos con el pensamiento por etapas de un abogado experimentado, combinado con la recuperación RAG de una base de conocimiento jurídico. Se activa cuando el usuario describe una disputa legal, consulta 'qué hago', 'puedo demandar', 'iré a la cárcel', pregunta por deudas impagas / ser demandado / pleitear / defensa de derechos / indemnización / accidentes laborales / accidentes de tráfico / conflictos de alquiler / sucesiones / divorcio / disputas contractuales / derechos del consumidor / conflictos laborales de empresa, pide analizar temas de contratos / agravios / laboral / familia / administrativo / participaciones societarias, o sube documentos legales. No procede para emitir dictámenes jurídicos formales, redactar documentos por encargo ni sustituir la comparecencia de un abogado. Admite la estrategia de lectura en capas, la búsqueda multi-rodada con cambio de términos, las listas de citas estructuradas (con estado de vigencia) y los modos rápido/experto. Incluye salvaguardas antialucinaciones (trazabilidad absoluta de fuentes + verificación de vigencia + lista de autocomprobación de alucinaciones) que garantizan citas normativas rastreables, jurisprudencia verificada y marcado automático de normas derogadas." }, "fr": { "name": "tencent-ima-copilot-legal-consultation", "summary": "Fournit un diagnostic préliminaire et des stratégies d'action pour les questions juridiques selon la démarche par étapes d'un avocat chevronné, combinée à la recherche RAG dans une base de connaissances juridiques. Déclenché lorsque l'utilisateur décrit un litige, demande 'que faire', 'puis-je poursuivre', 'risque-je la prison', s'interroge sur créances impayées / être poursuivi / aller en justice / défense des droits / indemnisation / accidents du travail / accidents de la route / litiges locatifs / successions / divorce / litiges contractuels / droits des consommateurs / litiges du travail en entreprise, demande l'analyse de questions contractuelles / délictuelles / de travail / familiales / administratives / de participations sociales, ou téléverse des documents juridiques. Ne convient pas pour émettre un avis juridique formel, rédiger des actes à la place d'un avocat ou le remplacer à l'audience. Prend en charge la stratégie de lecture en couches, la recherche multi-passes avec variation des termes, les listes de citations structurées (avec état de vigence) et les modes rapide/expert. Mécanismes anti-hallucination intégrés (traçabilité absolue des sources + vérification d'actualité + liste d'autocontrôle des hallucinations) : citations normatives traçables, jurisprudence vérifiée, marquage automatique des textes abrogés." }, "ja": { "name": "tencent-ima-copilot-legal-consultation", "summary": "ベテラン弁護士の段階的な事件処理の思考に沿い、RAG 法律ナレッジベース検索と組み合わせて、法律問題の初期診断と行動戦略を提供します。ユーザーが法的紛争を説明する、「どうすればいいか」「訴えられるか」「刑務所に入るか」と相談する、借金の未返済／訴えられる／訴訟を起こす／権利擁護／損害賠償／労災／交通事故／賃貸トラブル／遺産相続／離婚／契約紛争／消費者問題／会社の労務紛争などのシーンを尋ねる、契約／不法行為／労務／家事／行政／会社株式の問題の分析を求める、または法律文書をアップロードする場合に発動します。正式な法律意見書の発行、文書の代筆、弁護士出廷の代替には適しません。段階的精読戦略、複数ラウンドの語替え検索、構造化引用リスト（効力状態の注記付き）、クイック／エキスパートの 2 モードに対応。内蔵の反ハルシネーション保障機構（完全な出典追跡＋最新性検証＋ハルシネーション自己チェックリスト）により、条文引用の追跡可能性、判例の検証、失効条文の自動注記を確保します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, legalskill.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/legalskill/legalskill" }, { "id": "skill:elements-complaint-generator", "slug": "skill-elements-complaint-generator", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Use when converting a lawyer's already-written ordinary complaint (md/docx) or a conversational description into a Word document conforming to the Supreme People's Court's 67 official element-based complaint model text formats (Fa [2025] No. 82, rolled out nationwide on 2025-07-14). Applies to gener" }, "human": { "zh-CN": { "name": "elements-complaint-generator", "summary": "Use when converting 律师已写好的常规起诉状(md/docx)或对话描述为符合最高法 67 类官方要素式起诉状示范文本格式的 Word 文档(法〔2025〕82 号,2025-07-14 全国推广)。适用于民间借贷/离婚纠纷/机动车事故/劳动争议等所有 67 类民事/商事/行政/知产案由的要素式起诉状/答辩状生成。" }, "en": { "name": "elements-complaint-generator", "summary": "Use when converting a lawyer's already-written ordinary complaint (md/docx) or a conversational description into a Word document conforming to the Supreme People's Court's 67 official element-based complaint model text formats (Fa [2025] No. 82, rolled out nationwide on 2025-07-14). Applies to generating element-based complaints/statements of defense for all 67 civil/commercial/administrative/IP causes of action such as private lending, divorce disputes, motor vehicle accidents, and labor disputes." }, "es": { "name": "elements-complaint-generator", "summary": "Usar al convertir una demanda ordinaria ya redactada por el abogado (md/docx) o una descripción conversacional en un documento Word conforme a los 67 formatos oficiales de demanda por elementos del Tribunal Popular Supremo (Fa [2025] n.o 82, desplegado a nivel nacional el 14-07-2025). Aplica a la generación de demandas y contestaciones por elementos para las 67 causas civiles / comerciales / administrativas / de propiedad intelectual, como préstamos entre particulares, divorcios, accidentes de vehículos de motor y conflictos laborales." }, "fr": { "name": "elements-complaint-generator", "summary": "Utiliser pour convertir une requête ordinaire déjà rédigée par l'avocat (md/docx) ou une description conversationnelle en document Word conforme aux 67 formats officiels de requêtes à éléments du Tribunal populaire suprême (Fa [2025] n 82, déployé à l'échelle nationale le 14-07-2025). S'applique à la génération de requêtes et conclusions en défense à éléments pour les 67 causes civiles / commerciales / administratives / de propriété intellectuelle, telles que prêts entre particuliers, divorces, accidents de véhicules à moteur et litiges du travail." }, "ja": { "name": "elements-complaint-generator", "summary": "弁護士が既に書いた通常の訴状（md/docx）や会話での説明を、最高人民法院の 67 種類の公式要素式訴状モデルテキスト形式（法〔2025〕82 号、2025-07-14 に全国展開）に準拠した Word 文書へ変換する場合に使用します。民間貸借／離婚紛争／自動車事故／労務紛争など、民事・商事・行政・知財の全 67 類の請求原因に対応する要素式訴状・答弁状の生成に適用します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:patent-download", "slug": "skill-patent-download", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Batch patent PDF download tool. Use when the user needs to download full-text patent PDFs, query patent information, or bulk-export patent files. Supports multiple platforms (Google Patents preferred) and automatically handles application-number and publication-number formats." }, "human": { "zh-CN": { "name": "patent-download", "summary": "专利 PDF 批量下载工具。当用户需要下载专利全文 PDF、查询专利信息、批量导出专利文件时使用。支持多平台（Google Patents 优先），自动处理申请号和公告号格式。" }, "en": { "name": "patent-download", "summary": "Batch patent PDF download tool. Use when the user needs to download full-text patent PDFs, query patent information, or bulk-export patent files. Supports multiple platforms (Google Patents preferred) and automatically handles application-number and publication-number formats." }, "es": { "name": "patent-download", "summary": "Herramienta de descarga por lotes de PDF de patentes. Usar cuando el usuario necesite descargar PDF de texto completo de patentes, consultar información de patentes o exportar archivos de patentes en lote. Admite múltiples plataformas (Google Patents como preferente) y procesa automáticamente los formatos de número de solicitud y número de publicación." }, "fr": { "name": "patent-download", "summary": "Outil de téléchargement en lot de PDF de brevets. À utiliser lorsque l'utilisateur doit télécharger des PDF intégraux de brevets, interroger des informations de brevets ou exporter des fichiers de brevets en masse. Prend en charge plusieurs plateformes (Google Patents en priorité) et traite automatiquement les formats de numéro de demande et de numéro de publication." }, "ja": { "name": "patent-download", "summary": "特許 PDF の一括ダウンロードツール。ユーザーが特許の全文 PDF のダウンロード、特許情報の照会、特許ファイルの一括エクスポートを必要とする場合に使用します。複数プラットフォーム（Google Patents を優先）に対応し、出願番号と公告番号の形式を自動処理します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:yuandian-law-search", "slug": "skill-yuandian-law-search", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "A mechanism-aware legal research middleware layer for Yuandian retrieval. Use this skill when Chinese statutes or cases need to be queried, or when the user provides case facts, disputed issues, and existing legal analysis reports and asks for retrieval strategy design, favorable and adverse similar" }, "human": { "zh-CN": { "name": "yuandian-law-search", "summary": "元典检索机制感知型法律研究中间层。本技能应在需要查询中国法律法规或案例，或用户提供案件事实、争议焦点、既有法律分析报告并要求制定检索策略、查找正反类案、生成可追溯检索报告时使用；先做轻量案件研判与查询矩阵，再按向量、关键词和结构化字段调用元典 API 或 MCP。不要用于替代完整证据审查、诉讼方案或正式法律意见。" }, "en": { "name": "yuandian-law-search", "summary": "A mechanism-aware legal research middleware layer for Yuandian retrieval. Use this skill when Chinese statutes or cases need to be queried, or when the user provides case facts, disputed issues, and existing legal analysis reports and asks for retrieval strategy design, favorable and adverse similar-case lookup, or traceable retrieval reports; it first performs lightweight case assessment and a query matrix, then calls the Yuandian API or MCP via vectors, keywords, and structured fields. Not for: replacing full evidence review, litigation plans, or formal legal opinions." }, "es": { "name": "yuandian-law-search", "summary": "Capa intermedia de investigación jurídica consciente del mecanismo de búsqueda de Yuandian. Usar esta skill cuando sea necesario consultar leyes o casos chinos, o cuando el usuario aporte hechos del caso, puntos en disputa e informes de análisis jurídico previos y pida diseñar la estrategia de búsqueda, localizar casos análogos a favor y en contra, o generar informes de búsqueda trazables; primero realiza una valoración ligera del caso y una matriz de consultas, y luego llama a la API de Yuandian o a MCP mediante vectores, palabras clave y campos estructurados. No usar para: sustituir la revisión completa de pruebas, los planes litigiosos ni los dictámenes jurídicos formales." }, "fr": { "name": "yuandian-law-search", "summary": "Couche intermédiaire de recherche juridique consciente du mécanisme de recherche Yuandian. Utiliser cette compétence lorsqu'il faut interroger des lois ou affaires chinoises, ou lorsque l'utilisateur fournit des faits de l'affaire, des points litigieux et des rapports d'analyse juridique existants et demande de concevoir une stratégie de recherche, de trouver des jurisprudences analogues favorables et défavorables, ou de générer des rapports de recherche traçables ; elle réalise d'abord une appréciation légère de l'affaire et une matrice de requêtes, puis interroge l'API Yuandian ou MCP par vecteurs, mots-clés et champs structurés. Ne pas utiliser pour : remplacer l'examen complet des preuves, les plans contentieux ou les avis juridiques formels." }, "ja": { "name": "yuandian-law-search", "summary": "元典（Yuandian）検索メカニズムを認識するリーガルリサーチ中間層。中国の法令や判例の照会が必要な場合、またはユーザーが事件事実、争点、既存の法律分析報告書を提供し、検索戦略の策定、賛否両面の類似判例の検索、追跡可能な検索レポートの生成を求める場合に使用します。まず軽量な案件研判とクエリマトリクスを作成し、次にベクトル、キーワード、構造化フィールドで元典 API または MCP を呼び出します。完全な証拠審査、訴訟方案、正式な法律意見の代替には使用しないでください。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:law-to-markdown", "slug": "skill-law-to-markdown", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Converts statutes/regulatory documents (.txt/.docx/.pdf) to Markdown. Applies when the user asks for 'statutes to markdown' or 'pdf/docx to markdown'. When processing .pdf/.docx, first check whether the mineru-ocr skill is installed; if not installed, guide installation first, and after installation" }, "human": { "zh-CN": { "name": "law-to-markdown", "summary": "将法条/规范文件（.txt/.docx/.pdf）转为 Markdown。适用于用户要求“法条转 markdown”“pdf/docx 转 markdown”。处理 .pdf/.docx 时先检查是否已安装 mineru-ocr skill；未安装先引导安装，安装后优先用 mineru-ocr；仅在用户明确同意时再用本地回退方案。" }, "en": { "name": "law-to-markdown", "summary": "Converts statutes/regulatory documents (.txt/.docx/.pdf) to Markdown. Applies when the user asks for 'statutes to markdown' or 'pdf/docx to markdown'. When processing .pdf/.docx, first check whether the mineru-ocr skill is installed; if not installed, guide installation first, and after installation prefer mineru-ocr; use the local fallback only with the user's explicit consent." }, "es": { "name": "law-to-markdown", "summary": "Convierte normas/documentos normativos (.txt/.docx/.pdf) a Markdown. Aplica cuando el usuario pide 'convertir normas a markdown' o 'pdf/docx a markdown'. Al procesar .pdf/.docx, comprueba primero si está instalada la skill mineru-ocr; si no lo está, guía la instalación y, una vez instalada, prefiere mineru-ocr; usa la alternativa local solo con el consentimiento explícito del usuario." }, "fr": { "name": "law-to-markdown", "summary": "Convertit les textes de loi / documents normatifs (.txt/.docx/.pdf) en Markdown. S'applique lorsque l'utilisateur demande 'convertir des textes de loi en markdown' ou 'pdf/docx en markdown'. Lors du traitement de .pdf/.docx, vérifie d'abord si la compétence mineru-ocr est installée ; si elle ne l'est pas, guide d'abord l'installation, puis privilégie mineru-ocr une fois installée ; n'utilise la solution locale de repli qu'avec le consentement explicite de l'utilisateur." }, "ja": { "name": "law-to-markdown", "summary": "法令・規範文書（.txt/.docx/.pdf）を Markdown に変換します。「法令を markdown に」「pdf/docx を markdown に」という依頼に適用されます。.pdf/.docx の処理時は、まず mineru-ocr skill がインストール済みかを確認し、未インストールなら導入を案内し、導入後は mineru-ocr を優先します。ローカルのフォールバックはユーザーが明示的に同意した場合のみ使用します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:first-instance-criminal-defense", "slug": "skill-first-instance-criminal-defense", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Skill for first-instance criminal defense work, covering entrustment hand-over, supplementary case-file review, client meetings, pre-trial conferences, investigation and evidence gathering, exclusion of illegally obtained evidence, courtroom questioning/cross-examination/evidence presentation/argume" }, "human": { "zh-CN": { "name": "first-instance-criminal-defense", "summary": "刑事辩护一审阶段工作Skill，覆盖法院审判阶段的委托衔接、补充阅卷、会见、庭前会议、调查取证、非法证据排除、法庭发问质证举证辩论、认罪认罚、庭后工作与上诉评估。当用户提及\"一审阶段\"、\"刑事一审\"、\"开庭辩护\"、\"庭前会议\"、\"非法证据排除\"、\"刑事庭审\"等关键词时触发使用。" }, "en": { "name": "first-instance-criminal-defense", "summary": "Skill for first-instance criminal defense work, covering entrustment hand-over, supplementary case-file review, client meetings, pre-trial conferences, investigation and evidence gathering, exclusion of illegally obtained evidence, courtroom questioning/cross-examination/evidence presentation/argument, guilty plea and acceptance of punishment, post-trial work, and appeal assessment during the court trial stage. Triggered when the user mentions keywords such as 'first-instance stage', 'criminal first instance', 'trial defense', 'pre-trial conference', 'exclusion of illegal evidence', or 'criminal court hearing'." }, "es": { "name": "first-instance-criminal-defense", "summary": "Skill para el trabajo de defensa penal en primera instancia: cubre el enlace del encargo, la revisión complementaria del expediente, las entrevistas con el cliente, las conferencias previas al juicio, la investigación y obtención de pruebas, la exclusión de pruebas ilícitas, el interrogatorio / contrainterrogatorio / presentación de pruebas / alegatos en sala, la confesión y aceptación de sanción, el trabajo posjuicio y la valoración de la apelación durante la fase de juicio. Se activa cuando el usuario menciona palabras clave como 'primera instancia', 'primera instancia penal', 'defensa en juicio', 'conferencia previa', 'exclusión de pruebas ilícitas' o 'juicio oral penal'." }, "fr": { "name": "first-instance-criminal-defense", "summary": "Skill pour le travail de défense pénale en première instance : couvre la passation du mandat, la relecture complémentaire du dossier, les entretiens avec le client, les conférences préparatoires, l'enquête et l'obtention de preuves, l'exclusion des preuves illicites, l'interrogatoire / contre-interrogatoire / production des preuves / plaidoiries à l'audience, la reconnaissance de culpabilité et acceptation de la peine, le travail post-audience et l'évaluation de l'appel durant la phase de jugement. Déclenché lorsque l'utilisateur mentionne des mots-clés tels que 'première instance', 'première instance pénale', 'défense à l'audience', 'conférence préparatoire', 'exclusion des preuves illicites' ou 'audiences pénales'." }, "ja": { "name": "first-instance-criminal-defense", "summary": "刑事弁護の第一審段階の業務 skill。裁判審理段階における委任の引き継ぎ、補足の記録閲覧、接見、公判前会議、調査取証、違法収集証拠の排除、法廷での尋問・反対尋問・立証・弁論、認罪認罰、公判後の対応と控訴評価を網羅します。「第一審段階」「刑事第一審」「公判弁護」「公判前会議」「違法証拠排除」「刑事公判」などのキーワードにユーザーが言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:second-instance-criminal-defense", "slug": "skill-second-instance-criminal-defense", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Skill for second-instance criminal defense work, covering appeal-stage case intake, hand-over of first-instance materials, case-file review, client meeting strategy, seeking an oral hearing, written defense opinions, second-instance court hearings, communication for non-oral review, judgment analysi" }, "human": { "zh-CN": { "name": "second-instance-criminal-defense", "summary": "刑事辩护二审阶段工作Skill，覆盖上诉审委托接案、一审材料衔接、阅卷审查、会见策略、争取开庭、书面辩护意见、二审庭审、不开庭审理沟通、裁判分析及发回重审衔接。当用户提及\"二审\"、\"上诉\"、\"抗诉\"、\"争取开庭\"、\"发回重审\"、\"二审辩护\"等关键词时触发使用。" }, "en": { "name": "second-instance-criminal-defense", "summary": "Skill for second-instance criminal defense work, covering appeal-stage case intake, hand-over of first-instance materials, case-file review, client meeting strategy, seeking an oral hearing, written defense opinions, second-instance court hearings, communication for non-oral review, judgment analysis, and remand-for-retrial hand-off. Triggered when the user mentions keywords such as 'second instance', 'appeal', 'prosecutor's protest', 'seek an oral hearing', 'remand for retrial', or 'second-instance defense'." }, "es": { "name": "second-instance-criminal-defense", "summary": "Skill para el trabajo de defensa penal en segunda instancia: cubre la aceptación de casos en la fase de apelación, el enlace con los materiales de primera instancia, la revisión del expediente, la estrategia de entrevistas con el cliente, la solicitud de vista oral, los escritos de defensa, la vista de segunda instancia, la comunicación para la revisión sin vista oral, el análisis de la resolución y el enlace con el reenvío para nuevo juicio. Se activa cuando el usuario menciona palabras clave como 'segunda instancia', 'apelación', 'impugnación fiscal', 'solicitar vista oral', 'reenvío para nuevo juicio' o 'defensa en segunda instancia'." }, "fr": { "name": "second-instance-criminal-defense", "summary": "Skill pour le travail de défense pénale en seconde instance : couvre la prise en charge des dossiers en appel, la passation des pièces de première instance, l'examen du dossier, la stratégie d'entretiens avec le client, l'obtention d'une audience publique, les conclusions écrites de défense, l'audience de seconde instance, la communication pour un examen sans audience, l'analyse de la décision et le raccord au renvoi pour nouveau jugement. Déclenché lorsque l'utilisateur mentionne des mots-clés tels que 'seconde instance', 'appel', 'contre-appel du parquet', 'obtenir une audience', 'renvoi pour nouveau jugement' ou 'défense en seconde instance'." }, "ja": { "name": "second-instance-criminal-defense", "summary": "刑事弁護の第二審段階の業務 skill。控訴審での受任、第一審資料の引き継ぎ、記録閲覧・審査、接見戦略、公判開廷の獲得、書面弁護意見、第二審公判、開廷しない審理でのコミュニケーション、裁判の分析、差し戻し再審への接続を網羅します。「第二審」「控訴」「抗訴（検察による）」「開廷の獲得」「差し戻し再審」「第二審弁護」などのキーワードにユーザーが言及した場合に発動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:creditors-meeting", "slug": "skill-creditors-meeting", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "Systematically supports the full workflow of creditors' meetings — preparation, convening, voting, and minutes — covering three roles: administrator, creditors' agent, and debtor's agent. Supports the first and subsequent multiple creditors' meetings. Triggered when the user sends a client number an" }, "human": { "zh-CN": { "name": "creditors-meeting", "summary": "系统化支持债权人会议的筹备、召开、表决、纪要全流程工作，覆盖管理人、债权人代理、债务人代理三角色。支持第一次及后续多次债权人会议。当用户发送客户编号并提及\"债权人会议\"、\"召开会议\"、\"准备会议\"、\"会议通知\"、\"会议议程\"、\"会议纪要\"等指令时触发使用。" }, "en": { "name": "creditors-meeting", "summary": "Systematically supports the full workflow of creditors' meetings — preparation, convening, voting, and minutes — covering three roles: administrator, creditors' agent, and debtor's agent. Supports the first and subsequent multiple creditors' meetings. Triggered when the user sends a client number and mentions instructions such as \"creditors' meeting\", \"convene a meeting\", \"prepare a meeting\", \"meeting notice\", \"meeting agenda\", or \"meeting minutes\"." }, "es": { "name": "creditors-meeting", "summary": "Apoya de forma sistemática el flujo completo de la junta de acreedores —preparación, convocatoria, votación y acta—, cubriendo tres roles: administrador, agente de los acreedores y agente del deudor. Admite la primera junta y las múltiples juntas posteriores. Se activa cuando el usuario envía un número de cliente y menciona instrucciones como \"junta de acreedores\", \"convocar la junta\", \"preparar la junta\", \"aviso de convocatoria\", \"orden del día\" o \"acta de la junta\"." }, "fr": { "name": "creditors-meeting", "summary": "Prend en charge de manière systématique l'ensemble du flux de l'assemblée des créanciers — préparation, convocation, vote et procès-verbal — en couvrant trois rôles : administrateur, mandataire des créanciers et mandataire du débiteur. Prend en charge la première assemblée ainsi que les assemblées suivantes. Se déclenche lorsque l'utilisateur envoie un numéro de client et mentionne des instructions telles que « assemblée des créanciers », « convoquer l'assemblée », « préparer l'assemblée », « avis de convocation », « ordre du jour » ou « procès-verbal »." }, "ja": { "name": "creditors-meeting", "summary": "債権者集会の準備、招集、議決、議事録に至る全工程を体系的に支援し、管理人・債権者代理・債務者代理の3役割をカバーします。第1回目およびその後の複数回の債権者集会に対応します。ユーザーが顧客番号を送信し、「債権者集会」「集会の招集」「集会の準備」「集会通知」「議題」「議事録」などの指示に言及したときに起動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:creditor-claim-filing-review", "slug": "skill-creditor-claim-filing-review", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "A full-workflow Skill for bankruptcy claim management, covering claim filing, claim review, and preparation of the schedule of claims. Applicable to three roles: administrator (R1), creditors' agent (R2), and debtor's agent (R3). Triggered when the user sends a client number and mentions instruction" }, "human": { "zh-CN": { "name": "creditor-claim-filing-review", "summary": "破产债权管理全流程Skill，覆盖债权申报、审查、编制债权表全流程。适用于管理人（R1）、债权人代理（R2）、债务人代理（R3）三种角色。当用户发送客户编号并提及\"债权申报\"、\"债权审查\"、\"编制债权表\"、\"申报债权\"等指令时触发使用。" }, "en": { "name": "creditor-claim-filing-review", "summary": "A full-workflow Skill for bankruptcy claim management, covering claim filing, claim review, and preparation of the schedule of claims. Applicable to three roles: administrator (R1), creditors' agent (R2), and debtor's agent (R3). Triggered when the user sends a client number and mentions instructions such as \"claim filing\", \"claim review\", \"prepare the schedule of claims\", or \"file a claim\"." }, "es": { "name": "creditor-claim-filing-review", "summary": "Un Skill de flujo completo para la gestión de créditos concursales, que cubre la presentación de créditos, su revisión y la elaboración del cuadro de créditos. Aplicable a tres roles: administrador (R1), agente de los acreedores (R2) y agente del deudor (R3). Se activa cuando el usuario envía un número de cliente y menciona instrucciones como \"presentación de créditos\", \"revisión de créditos\", \"elaborar el cuadro de créditos\" o \"presentar un crédito\"." }, "fr": { "name": "creditor-claim-filing-review", "summary": "Une Skill couvrant l'ensemble du flux de gestion des créances de la faillite : dépôt des créances, examen des créances et établissement du tableau des créances. Applicable à trois rôles : administrateur (R1), mandataire des créanciers (R2) et mandataire du débiteur (R3). Se déclenche lorsque l'utilisateur envoie un numéro de client et mentionne des instructions telles que « dépôt de créance », « examen des créances », « établir le tableau des créances » ou « déclarer une créance »." }, "ja": { "name": "creditor-claim-filing-review", "summary": "破産債権管理の全プロセスをカバーするSkillで、債権届出、債権審査、債権表の作成の全流れを扱います。管理人（R1）、債権者代理（R2）、債務者代理（R3）の3役割に対応します。ユーザーが顧客番号を送信し、「債権届出」「債権審査」「債権表の作成」「債権の届出」などの指示に言及したときに起動します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:legalskill-legal-consultation", "slug": "skill-legalskill-legal-consultation", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "legalskill" ], "canonical": { "brief": "Provides lawyers and in-house counsel with statute retrieval, regulation lookup, case study, precedent search, guiding-case lookup, judicial-interpretation analysis, contract review, and professional assessment; provides the general public with legal consultation, case analysis, and rights-protectio" }, "human": { "zh-CN": { "name": "legalskill-legal-consultation", "summary": "为律师、法务提供法条检索、法规查询、案例研究、判例检索、指导案例查找、司法解释解读、合同审查与专业研判，为普通公众提供法律咨询、案件分析和维权指导。适用于普通用户描述日常纠纷、追问法律后果或寻求行动策略，即使用户未使用法律术语，常见说法如欠钱不还、被起诉、打官司、合同违约、离婚、抚养权、抚养费、探视权、家暴、彩礼、遗产继承、劳动纠纷、工伤、加班费、年终奖、竞业限制、违法解除、违法辞退、裁员、被辞退、经济补偿、年休假、交通事故、车祸、租房押金不退、消费者投诉、公司股权争议、合伙纠纷、股权代持、股东知情权、拆迁补偿、行政处罚、行政复议、知识产权侵权、商标侵权、专利纠纷、著作权纠纷、商业秘密、不正当竞争、房屋买卖纠纷、物业纠纷、邻里纠纷、名誉权侵权、隐私泄露、网络暴力、医疗事故、医疗纠纷、保险合同纠纷、人身损害、被人打了、食品安全、产品责任等，或问\"怎么办\"\"能不能告\"\"会不会坐牢\"\"能赔多少\"\"胜算大吗\"\"有什么风险\"\"要赔多少钱\"\"会不会被拘留\"\"会不会判刑\"\"能不能缓刑\"\"怎么应诉\"\"如何反诉\"\"如何上诉\"\"打官司要多久\"\"证据不足怎么办\"\"对方不履行怎么办\"\"律师费多少\"\"仲裁和诉讼哪个好\"\"要不要请律师\"\"自己辩护行不行\"\"调解书有法律效力吗\"\"强制执行怎么申请\"\"被列入失信名单怎么办\"\"被当成老赖怎么办\"\"对方转移财产怎么办\"\"判决后不执行怎么办\"\"取保候审要什么条件\"\"能不能不起诉\"等。也覆盖律师、法务的高频专业任务：按案号查案例、按裁判要旨找类案、法条沿革追溯、法条效力核实、合同条款分析与漏洞审查、违约赔偿计算、劳动关系认定、竞业限制审查、裁判口径对比与各地裁判标准差异分析、合规审查、诉讼风险评估、仲裁策略设计、调解方案拟订、强制执行与财产保全分析、刑事报案可行性初评、证据链梳理、庭审准备提纲、再审申请、执行异议、破产债权申报等。用户上传合同、判决书、起诉状、答辩状、仲裁申请书、执行申请书、律师函等法律文件时自动触发分析，支持多轮追问与策略细化。不适用于出具正式法律意见书、代写法律文书、代理出庭、学术写作或非中文法律检索。" }, "en": { "name": "legalskill-legal-consultation", "summary": "Provides lawyers and in-house counsel with statute retrieval, regulation lookup, case study, precedent search, guiding-case lookup, judicial-interpretation analysis, contract review, and professional assessment; provides the general public with legal consultation, case analysis, and rights-protection guidance. Suits ordinary users describing everyday disputes, asking about legal consequences, or seeking action strategies, even when they use no legal terminology. Common phrasings include: owing money and not repaying, being sued, going to court, breach of contract, divorce, custody, child support, visitation rights, domestic violence, betrothal gifts (caili), inheritance, labor disputes, work injuries, overtime pay, year-end bonus, non-compete restrictions, unlawful termination, unlawful dismissal, layoffs, being dismissed, severance compensation, paid annual leave, traffic accidents, car crashes, unreturned rental deposits, consumer complaints, corporate equity disputes, partnership disputes, nominee shareholding, shareholders' right to information, demolition compensation, administrative penalties, administrative reconsideration, IP infringement, trademark infringement, patent disputes, copyright disputes, trade secrets, unfair competition, property sale disputes, property management disputes, neighborhood disputes, reputation infringement, privacy leaks, cyberbullying, medical malpractice, medical disputes, insurance contract disputes, personal injury, being assaulted, food safety, product liability, and the like; or questions such as \"What should I do?\", \"Can I sue?\", \"Will I go to jail?\", \"How much compensation can I get?\", \"Are my chances good?\", \"What are the risks?\", \"How much do I have to pay?\", \"Will I be detained?\", \"Will I be sentenced?\", \"Can I get probation?\", \"How do I respond to the suit?\", \"How do I counterclaim?\", \"How do I appeal?\", \"How long does a lawsuit take?\", \"What if the evidence is insufficient?\", \"What if the other party does not perform?\", \"How much are attorney fees?\", \"Arbitration or litigation — which is better?\", \"Do I need a lawyer?\", \"Can I defend myself?\", \"Is a mediation agreement legally binding?\", \"How do I apply for compulsory enforcement?\", \"What if I am placed on the dishonesty list?\", \"What if I am treated as a deadbeat debtor?\", \"What if the other party transfers assets away?\", \"What if the judgment is not enforced?\", \"What are the conditions for release on bail pending trial?\", \"Can there be no prosecution?\", and the like. It also covers the high-frequency professional tasks of lawyers and in-house counsel: finding cases by case number, finding similar cases by holding, tracing the revision history of provisions, verifying provision validity, contract clause analysis and loophole review, breach-damages calculation, labor-relationship determination, non-compete review, comparison of adjudication approaches and analysis of regional differences in adjudication standards, compliance review, litigation risk assessment, arbitration strategy design, drafting mediation plans, enforcement and asset-preservation analysis, preliminary feasibility assessment of criminal complaints, evidence-chain mapping, trial-preparation outlines, retrial applications, enforcement objections, bankruptcy claim filing, and more. Analysis is triggered automatically when the user uploads legal documents such as contracts, judgments, complaints, statements of defense, arbitration applications, enforcement applications, or attorney letters; multi-turn follow-up questions and strategy refinement are supported. Not suitable for issuing formal legal opinions, ghost-writing legal documents, court representation, academic writing, or non-Chinese legal research." }, "es": { "name": "legalskill-legal-consultation", "summary": "Ofrece a abogados y equipos jurídicos búsqueda de artículos, consulta de normas, estudio de casos, búsqueda de precedentes, localización de casos guía, análisis de interpretaciones judiciales, revisión de contratos y valoración profesional; ofrece al público general asesoría legal, análisis de casos y orientación para la defensa de derechos. Apto para usuarios comunes que describen disputas cotidianas, preguntan por las consecuencias legales o buscan estrategias de actuación, aunque no usen terminología jurídica. Formulaciones habituales: deudas impagas, ser demandado, pleitear, incumplimiento de contrato, divorcio, custodia, pensión de alimentos, régimen de visitas, violencia doméstica, regalos de esponsales (caili), herencia, disputas laborales, accidentes de trabajo, pago de horas extra, aguinaldo, pactos de no competencia, terminación ilegal del contrato, despido ilegal, despidos colectivos, ser despedido, indemnización económica, vacaciones anuales pagadas, accidentes de tráfico, choques de coche, depósitos de alquiler no devueltos, reclamaciones de consumo, disputas sobre participaciones societarias, disputas entre socios, tenencia de acciones a nombre de terceros, derecho de información del accionista, indemnización por demolición, sanciones administrativas, revisión administrativa, infracción de propiedad intelectual, infracción de marca, disputas de patentes, disputas de derechos de autor, secretos empresariales, competencia desleal, disputas de compraventa de vivienda, disputas de administración de fincas, disputas vecinales, infracción del derecho al honor, filtraciones de privacidad, acoso en línea, negligencia médica, disputas médicas, disputas de contratos de seguro, daños personales, agresiones, seguridad alimentaria, responsabilidad del producto, etc.; o preguntas como «¿Qué hago?», «¿Puedo demandar?», «¿Iré a la cárcel?», «¿Cuánta indemnización puedo obtener?», «¿Tengo buenas opciones?», «¿Qué riesgos hay?», «¿Cuánto debo pagar?», «¿Me detendrán?», «¿Me condenarán?», «¿Puedo obtener la suspensión de la pena?», «¿Cómo contesto a la demanda?», «¿Cómo presento una reconvención?», «¿Cómo apelo?», «¿Cuánto dura un pleito?», «¿Qué hago si faltan pruebas?», «¿Qué hago si la otra parte no cumple?», «¿Cuánto cuestan los honorarios?», «¿Arbitraje o litigio, qué es mejor?», «¿Necesito abogado?», «¿Puedo defenderme yo mismo?», «¿El acta de mediación tiene efectos jurídicos?», «¿Cómo solicito la ejecución forzosa?», «¿Qué hago si figuro en la lista de incumplidos?», «¿Qué hago si me tratan como deudor moroso?», «¿Qué hago si la otra parte traslada sus bienes?», «¿Qué hago si la sentencia no se ejecuta?», «¿Qué requisitos exige la libertad bajo fianza pendiente de juicio?», «¿Es posible no procesar?», etc. También cubre las tareas profesionales de alta frecuencia de abogados y equipos jurídicos: buscar casos por número de expediente, hallar casos análogos por doctrina esencial, rastrear la evolución de los artículos, verificar la vigencia de los artículos, análisis de cláusulas y revisión de lagunas contractuales, cálculo de indemnización por incumplimiento, determinación de la relación laboral, revisión de pactos de no competencia, comparación de criterios de decisión y análisis de diferencias regionales en los estándares de decisión, revisión de cumplimiento, evaluación de riesgos litigiosos, diseño de estrategias de arbitraje, elaboración de propuestas de mediación, análisis de ejecución forzosa y medidas cautelares, valoración preliminar de la viabilidad de una denuncia penal, ordenación de la cadena probatoria, esquemas de preparación del juicio, solicitudes de revisión, objeciones a la ejecución, presentación de créditos concursales, etc. El análisis se activa automáticamente cuando el usuario sube documentos jurídicos como contratos, sentencias, demandas, contestaciones, solicitudes de arbitraje, solicitudes de ejecución o cartas de abogado; admite preguntas de seguimiento en varias rondas y el refinamiento de la estrategia. No es apto para emitir dictámenes jurídicos formales, redactar documentos jurídicos por encargo, representación en juicio, redacción académica ni investigación jurídica en idiomas no chinos." }, "fr": { "name": "legalskill-legal-consultation", "summary": "Offre aux avocats et aux équipes juridiques la recherche d'articles de loi, la consultation de textes réglementaires, l'étude de cas, la recherche de jurisprudence, la recherche d'affaires directrices, l'analyse des interprétations judiciaires, la revue de contrats et l'expertise professionnelle ; offre au grand public consultation juridique, analyse de dossiers et accompagnement dans la défense des droits. Convient aux utilisateurs ordinaires qui décrivent des litiges du quotidien, s'interrogent sur les conséquences juridiques ou cherchent des stratégies d'action, même sans vocabulaire juridique. Formulations courantes : dettes impayées, être poursuivi en justice, aller au procès, inexécution contractuelle, divorce, garde d'enfant, pension alimentaire, droit de visite, violences conjugales, cadeaux de fiançailles (caili), succession, litiges du travail, accidents du travail, heures supplémentaires, prime de fin d'année, clauses de non-concurrence, résiliation illégale du contrat, licenciement illégal, plans de licenciement, être licencié, indemnité de départ, congés annuels payés, accidents de la circulation, collisions de véhicules, caution de location non restituée, réclamations de consommateurs, litiges sur les participations sociétaires, différends entre associés, détention d'actions pour le compte d'autrui, droit à l'information de l'actionnaire, indemnisation de démolition, sanctions administratives, recours administratif, atteinte à la propriété intellectuelle, contrefaçon de marque, litiges de brevets, litiges de droit d'auteur, secrets d'affaires, concurrence déloyale, litiges de vente immobilière, litiges de gestion d'immeubles, différends de voisinage, atteinte à la réputation, fuites de vie privée, cyberviolence, accidents médicaux, litiges médicaux, litiges de contrats d'assurance, dommages corporels, voies de fait, sécurité alimentaire, responsabilité du produit, etc. ; ou des questions telles que « Que faire ? », « Puis-je poursuivre ? », « Vais-je en prison ? », « Quelle indemnisation puis-je obtenir ? », « Ai-je de bonnes chances ? », « Quels sont les risques ? », « Combien dois-je payer ? », « Serai-je détenu ? », « Serai-je condamné ? », « Puis-je obtenir un sursis ? », « Comment répondre à la poursuite ? », « Comment former une demande reconventionnelle ? », « Comment faire appel ? », « Combien de temps dure un procès ? », « Que faire si les preuves sont insuffisantes ? », « Que faire si l'autre partie n'exécute pas ? », « Combien coûtent les honoraires d'avocat ? », « Arbitrage ou contentieux, que choisir ? », « Faut-il prendre un avocat ? », « Puis-je me défendre moi-même ? », « Le procès-verbal de médiation a-t-il force juridique ? », « Comment demander l'exécution forcée ? », « Que faire si je figure sur la liste des débiteurs défaillants ? », « Que faire si l'on me traite de mauvais payeur ? », « Que faire si l'autre partie transfère ses actifs ? », « Que faire si le jugement n'est pas exécuté ? », « Quelles conditions pour la remise en liberté sous caution dans l'attente du procès ? », « Peut-on renoncer aux poursuites ? », etc. Elle couvre aussi les tâches professionnelles à haute fréquence des avocats et équipes juridiques : trouver des affaires par numéro de dossier, trouver des affaires analogues par motifs essentiels, retracer l'historique des révisions des dispositions, vérifier la validité des dispositions, analyse des clauses et revue des lacunes contractuelles, calcul des indemnités d'inexécution, qualification de la relation de travail, revue des clauses de non-concurrence, comparaison des approches jurisprudentielles et analyse des écarts régionaux de standards de jugement, revue de conformité, évaluation des risques contentieux, conception de stratégies d'arbitrage, élaboration de propositions de médiation, analyse de l'exécution forcée et des mesures conservatoires, évaluation préliminaire de la faisabilité d'une plainte pénale, structuration de la chaîne de preuves, plans de préparation de l'audience, demandes en révision, contestations de l'exécution, déclaration de créances faillitaires, etc. L'analyse se déclenche automatiquement lorsque l'utilisateur téléverse des documents juridiques tels que contrats, jugements, conclusions de poursuite, conclusions en défense, requêtes d'arbitrage, requêtes d'exécution ou lettres d'avocat ; les questions de suivi en plusieurs tours et le raffinement de la stratégie sont pris en charge. Ne convient pas pour : l'émission d'avis juridiques formels, la rédaction de documents juridiques pour autrui, la représentation à l'audience, la rédaction académique ou la recherche juridique hors de la langue chinoise." }, "ja": { "name": "legalskill-legal-consultation", "summary": "弁護士・法務担当者には条文検索、法令照会、事例研究、判例検索、指導案例の検索、司法解釈の解読、契約審査、専門的な検討を提供し、一般の人々には法律相談、事件分析、権利保護のガイダンスを提供します。法律用語を使わない一般ユーザーが日常的な紛争を説明し、法的帰結を尋ねたり行動戦略を求めたりする場合に適しています。よくある表現：お金を貸したのに返さない、訴えられた、裁判を起こす、契約違反、離婚、親権、養育費、面会交流権、DV、結納金（彩礼）、相続、労務紛争、労災、残業代、年末賞与、競業避止、違法な契約解除、違法解雇、リストラ、解雇された、経済補償、年次有給休暇、交通事故、自動車事故、賃貸保証金の不返還、消費者クレーム、会社の株式をめぐる争い、パートナーシップ紛争、名義株（株式の名義預かり）、株主の知情権、立ち退き補償、行政処分、行政不服申立て、知的財産権侵害、商標侵害、特許紛争、著作権紛争、営業秘密、不正競争、住宅売買紛争、物件管理紛争、近隣紛争、名誉権侵害、プライバシーの漏洩、ネットいじめ、医療事故、医療紛争、保険契約紛争、人身損害、暴行を受けた、食品安全、製品責任など。または「どうすればいいか」「訴えられるか」「刑務所に入るか」「賠償はいくらもらえるか」「勝ち目はあるか」「どんなリスクがあるか」「いくら賠償しなければならないか」「拘留されるか」「刑を言い渡されるか」「執行猶予は得られるか」「どう答弁するか」「反訴の方法」「控訴の方法」「訴訟はどれくらいかかるか」「証拠が不十分な場合は」「相手が履行しない場合は」「弁護士費用はいくらか」「仲裁と訴訟どちらがいいか」「弁護士に依頼すべきか」「自分で弁護できるか」「調停調書に法的効力はあるか」「強制執行の申立て方法」「失信名簿に載った場合は」「債務不履行者（老頼）扱いされた場合は」「相手が財産を移転した場合は」「判決後も履行しない場合は」「保釈（取保候審）の条件は」「不起訴にできるか」など。弁護士・法務の高頻度の専門タスクもカバーします：事件番号による事例検索、裁判要旨による類似事件検索、条文の沿革追跡、条文の効力確認、契約条項の分析と不備審査、違約賠償の計算、労働関係の認定、競業避止の審査、裁判基準の比較と各地の裁判基準差の分析、コンプライアンス審査、訴訟リスク評価、仲裁戦略の設計、調停案の作成、強制執行と財産保全の分析、刑事告訴の実行可能性の予備評価、証拠連鎖の整理、公判準備のアウトライン、再審申立て、執行異議、破産債権の届出など。ユーザーが契約書、判決書、訴状、答弁書、仲裁申立書、執行申立書、弁護士警告書などの法律文書をアップロードすると自動的に分析を起動し、複数回の追及質問と戦略の具体化に対応します。正式な法律意見書の発行、法律文書の代筆、出廷代理、学術執筆、中国語以外の法律検索には適しません。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, legalskill.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/legalskill/legalskill" }, { "id": "skill:criminal-defense-coordinator", "slug": "skill-criminal-defense-coordinator", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "pa1nrui1" ], "canonical": { "brief": "A criminal-defense workflow orchestration Skill that performs professional routing based on criminal-procedure stages and special tags, stage-transition prompts, key-deadline reminders, and next-step recommendations. Triggered when the user mentions keywords such as \"criminal case\", \"criminal defens" }, "human": { "zh-CN": { "name": "criminal-defense-coordinator", "summary": "刑事辩护专业流程编排 Skill，负责根据刑事诉讼阶段和特殊标签进行专业路由、 阶段转换提示、关键期限提示和下一步工作建议。当用户提及\"刑事案件\"、 \"刑事辩护\"、\"刑辩\"等关键词，或需要跨阶段协调刑事辩护工作时触发。 事项建档、路径、文件读取、来源披露、OCR 校正和缺口归档由「法律工作总控」统一处理。" }, "en": { "name": "criminal-defense-coordinator", "summary": "A criminal-defense workflow orchestration Skill that performs professional routing based on criminal-procedure stages and special tags, stage-transition prompts, key-deadline reminders, and next-step recommendations. Triggered when the user mentions keywords such as \"criminal case\", \"criminal defense\", or \"xingbian (criminal defense)\", or needs to coordinate criminal-defense work across stages. Case filing, routing, file reading, source disclosure, OCR correction, and gap archiving are handled centrally by the \"Legal Work Master Control\"." }, "es": { "name": "criminal-defense-coordinator", "summary": "Un Skill de orquestación del flujo de defensa penal que realiza el enrutamiento profesional según las etapas del proceso penal y etiquetas especiales, avisos de transición de etapa, recordatorios de plazos clave y sugerencias de siguientes pasos. Se activa cuando el usuario menciona palabras clave como \"caso penal\", \"defensa penal\" o \"penal\", o necesita coordinar el trabajo de defensa penal entre etapas. La apertura del expediente, el enrutamiento, la lectura de archivos, la divulgación de fuentes, la corrección OCR y el archivo de brechas se gestionan de forma centralizada en el «Control maestro de trabajo legal»." }, "fr": { "name": "criminal-defense-coordinator", "summary": "Une Skill d'orchestration du flux de défense pénale : routage professionnel selon les phases de la procédure pénale et les étiquettes spéciales, alertes de transition de phase, rappels des délais clés et suggestions d'étapes suivantes. Se déclenche lorsque l'utilisateur mentionne des mots-clés tels que « affaire pénale », « défense pénale » ou « pénal », ou doit coordonner le travail de défense pénale entre plusieurs phases. L'ouverture du dossier, le routage, la lecture des fichiers, la divulgation des sources, la correction OCR et l'archivage des lacunes sont gérés de manière centralisée par le « Contrôle général du travail juridique »." }, "ja": { "name": "criminal-defense-coordinator", "summary": "刑事弁護の業務フローを編成するSkillです。刑事手続の段階や特殊タグに基づく専門ルーティング、段階移行の提示、重要期限のリマインド、次のステップの提案を担当します。ユーザーが「刑事事件」「刑事弁護」「刑弁」などのキーワードに言及した場合や、段階をまたぐ刑事弁護業務の調整が必要なときに起動します。事件の記録作成、経路、ファイル読み取り、出所の開示、OCR修正、欠落のアーカイブは「法律業務総控（マスターコントロール）」が一括して処理します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, pa1nrui1.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/pa1nrui1/legal-skills" }, { "id": "skill:legal-industry-report", "slug": "skill-legal-industry-report", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "cat-xierluo" ], "canonical": { "brief": "Takes an industry as input (with optional region and business focus) and produces a deliverable, finely typeset A4 PDF industry legal research report. The design follows the law-firm blue-paper style (dark blue as the primary color + white pages + large serif headlines + top brand banner), includes " }, "human": { "zh-CN": { "name": "legal-industry-report", "summary": "输入一个行业（可选区域与业务方向），输出一份可交付的精排 A4 PDF 行业法律调研报告。设计风格采用律所蓝皮书体例（深蓝主色 + 白色页面 + 衬线大标题 + 顶部品牌带），内置 report-profile.md 个性化配置（律所抬头 / 配色 / 主办律师 / 封面变体 / 联系方式），数据源来自企查查 MCP（企业名录 + 司法/经营风险标签）+ 网络检索（政策文件、产业园名录、产业链结构）。建议在用户说\"做一份行业调研报告 / 见客户前先了解下 XX 行业 / 这周要去 XX 行业跑客户帮我准备背景资料 / 开拓新行业之前知己知彼\"时使用；不要用于：客户名下的具体法律意见、对单一企业的尽调（用 opc-legal-counsel）、按周推送的对外研报（用 weekly-legal-briefing）。" }, "en": { "name": "legal-industry-report", "summary": "Takes an industry as input (with optional region and business focus) and produces a deliverable, finely typeset A4 PDF industry legal research report. The design follows the law-firm blue-paper style (dark blue as the primary color + white pages + large serif headlines + top brand banner), includes a built-in report-profile.md personalization config (law-firm letterhead / color scheme / lead attorney / cover variants / contact details), and draws data from Qichacha MCP (company directory + judicial/operational risk tags) plus web search (policy documents, industrial park directories, industry chain structure). Recommended when the user says \"prepare an industry research report / help me get to know industry XX before meeting a client / I'll be visiting clients in industry XX this week, prepare background material for me / know yourself and the other side before expanding into a new industry\". Do not use for: specific legal opinions under a client's matter, due diligence on a single company (use opc-legal-counsel), or externally distributed research pushed weekly (use weekly-legal-briefing)." }, "es": { "name": "legal-industry-report", "summary": "Toma como entrada un sector (con región y enfoque de negocio opcionales) y produce un informe de investigación jurídica sectorial en PDF A4, finamente maquetado y entregable. El diseño sigue el estilo de libro azul de los despachos (azul oscuro como color principal + páginas blancas + titulares grandes con serif + banda de marca superior), incluye una configuración de personalización integrada en report-profile.md (membrete del despacho / paleta de colores / abogado responsable / variantes de portada / datos de contacto) y obtiene los datos de Qichacha MCP (directorio de empresas + etiquetas de riesgo judicial/operativo) más búsqueda web (documentos de políticas, directorios de parques industriales, estructura de la cadena industrial). Se recomienda cuando el usuario dice \"prepárame un informe de investigación sectorial / antes de ver al cliente quiero conocer el sector XX / esta semana visito clientes del sector XX, prepara material de contexto / antes de abrir un sector nuevo, conócete a ti mismo y conoce a tu contraparte\". No usar para: opiniones jurídicas específicas de un asunto del cliente, due diligence de una sola empresa (usar opc-legal-counsel) ni informes externos con envío semanal (usar weekly-legal-briefing)." }, "fr": { "name": "legal-industry-report", "summary": "Prend en entrée un secteur d'activité (région et orientation métier facultatives) et produit un rapport de recherche juridique sectoriel A4 en PDF, finement mis en page et livrable. Le design reprend le style « livre bleu » des cabinets (bleu foncé comme couleur principale + pages blanches + grands titres en serif + bandeau de marque en haut), intègre une configuration de personnalisation via report-profile.md (en-tête du cabinet / palette de couleurs / avocat en charge / variantes de couverture / coordonnées) et s'appuie sur Qichacha MCP (annuaire d'entreprises + étiquettes de risque judiciaire/exploitation) plus la recherche web (documents de politique publique, annuaires de parcs industriels, structure des chaînes industrielles). Recommandé lorsque l'utilisateur dit « prépare un rapport d'étude sectorielle / avant de voir le client, fais-moi connaître le secteur XX / cette semaine je rends visite à des clients du secteur XX, prépare-moi la documentation de fond / avant d'ouvrir un nouveau secteur, connais-toi toi-même et connais ton adversaire ». À ne pas utiliser pour : des avis juridiques spécifiques au dossier d'un client, la due diligence d'une seule entreprise (utiliser opc-legal-counsel) ou une publication externe hebdomadaire (utiliser weekly-legal-briefing)." }, "ja": { "name": "legal-industry-report", "summary": "業界を1つ入力すると（地域と業務領域は任意）、納品可能な整形済みA4 PDFの業界法務調査レポートを出力します。デザインは法律事務所のブルーブック形式（濃紺ベース＋白ページ＋セリフ体の大見出し＋上部のブランド帯）を採用し、report-profile.md によるカスタマイズ設定（事務所のレターヘッド／配色／担当弁護士／表紙バリエーション／連絡先）を内蔵し、データソースは Qichacha（企查查）MCP（企業リスト＋司法・経営リスクタグ）＋ウェブ検索（政策文書、産業園区リスト、産業チェーン構造）です。ユーザーが「業界調査レポートを作って／顧客に会う前にXX業界を把握したい／今週XX業界の顧客を回るので背景資料を準備して／新業界開拓の前に敵を知り己を知れ」と言ったときの使用を推奨します。次の用途には使わないでください：特定顧客の案件に関する具体的な法律意見、単一企業のデューデリジェンス（opc-legal-counsel を使用）、週次で配信する対外リサーチレポート（weekly-legal-briefing を使用）。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, cat-xierluo.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/cat-xierluo/legal-skills" }, { "id": "skill:legal-abductive-reasoning", "slug": "skill-legal-abductive-reasoning", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "A legal abductive-reasoning skill for generating and evaluating the most plausible explanatory hypotheses when evidence is incomplete and facts are ambiguous. Combines Mill's five methods (agreement, difference, joint method of agreement and difference, residues, concomitant variation) for structure" }, "human": { "zh-CN": { "name": "legal-abductive-reasoning", "summary": "法律溯因推理（Abductive Reasoning）技能，用于在证据不完整、事实模糊时生成并评估最合理的解释性假设。结合米尔五法（求同、求异、求同求异、剩余、共变）进行结构化因果推断。当用户需要\"提出可能的法律解释\"、\"生成辩护/起诉假设\"、\"评估证据的多种解读\"或\"基于不完整事实推断最佳解释\"时触发。适用于案例分析、证据评估、规则适用争议、因果关系推断等场景。" }, "en": { "name": "legal-abductive-reasoning", "summary": "A legal abductive-reasoning skill for generating and evaluating the most plausible explanatory hypotheses when evidence is incomplete and facts are ambiguous. Combines Mill's five methods (agreement, difference, joint method of agreement and difference, residues, concomitant variation) for structured causal inference. Triggered when the user needs to \"propose possible legal explanations\", \"generate defense/prosecution hypotheses\", \"evaluate multiple readings of evidence\", or \"infer the best explanation from incomplete facts\". Suits case analysis, evidence assessment, disputes over rule application, causal inference, and similar scenarios." }, "es": { "name": "legal-abductive-reasoning", "summary": "Habilidad de razonamiento abductivo jurídico para generar y evaluar las hipótesis explicativas más plausibles cuando la evidencia es incompleta y los hechos ambiguos. Combina los cinco métodos de Mill (concordancia, diferencia, método combinado de concordancia y diferencia, residuos, variación concomitante) para una inferencia causal estructurada. Se activa cuando el usuario necesita \"proponer posibles explicaciones jurídicas\", \"generar hipótesis de defensa/acusación\", \"evaluar múltiples lecturas de la evidencia\" o \"inferir la mejor explicación a partir de hechos incompletos\". Apto para análisis de casos, evaluación de pruebas, disputas sobre la aplicación de normas, inferencia causal y escenarios similares." }, "fr": { "name": "legal-abductive-reasoning", "summary": "Compétence de raisonnement abductif juridique, destinée à générer et évaluer les hypothèses explicatives les plus plausibles lorsque les preuves sont incomplètes et les faits ambigus. Combine les cinq méthodes de Mill (concordance, différence, méthode conjointe de concordance et de différence, résidus, variation concomitante) pour une inférence causale structurée. Se déclenche lorsque l'utilisateur doit « proposer des explications juridiques possibles », « générer des hypothèses de défense/poursuite », « évaluer plusieurs lectures des preuves » ou « déduire la meilleure explication à partir de faits incomplets ». Convient à l'analyse de dossiers, l'évaluation des preuves, les contestations d'application de règles, l'inférence causale et d'autres scénarios." }, "ja": { "name": "legal-abductive-reasoning", "summary": "法律アブダクション（溯因推理）のスキルです。証拠が不完全で事実が曖昧な場合に、最も尤もらしい説明仮説を生成・評価します。ミルの五法（一致法、差異法、一致差異併用法、残余法、共変法）を組み合わせて構造化された因果推論を行います。「あり得る法的説明を出して」「弁護/起訴の仮説を作って」「証拠の複数の解釈を評価して」「不完全な事実から最善の説明を推論して」というニーズがあるときに起動します。ケース分析、証拠評価、規範適用の争点、因果関係の推論などの場面に適しています。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:normative-meaning-argumentation", "slug": "skill-normative-meaning-argumentation", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Analyzes the purpose and value orientation of legal norms, and how facts should enter normative evaluation, determining the limits of subsumption; must be appropriately distinguished from teleological (normative-purpose) interpretation within legal interpretation." }, "human": { "zh-CN": { "name": "normative-meaning-argumentation", "summary": "分析法律规范的目的、价值导向，以及事实应如何进入规范评价，判断涵摄的限度，需要适当区别于法律解释中的规范目的解释。" }, "en": { "name": "normative-meaning-argumentation", "summary": "Analyzes the purpose and value orientation of legal norms, and how facts should enter normative evaluation, determining the limits of subsumption; must be appropriately distinguished from teleological (normative-purpose) interpretation within legal interpretation." }, "es": { "name": "normative-meaning-argumentation", "summary": "Analiza el propósito y la orientación axiológica de las normas jurídicas, así como cómo deben incorporarse los hechos a la valoración normativa, y determina los límites de la subsunción; debe distinguirse apropiadamente de la interpretación teleológica (según el propósito normativo) dentro de la interpretación jurídica." }, "fr": { "name": "normative-meaning-argumentation", "summary": "Analyse la finalité et l'orientation axiologique des normes juridiques, ainsi que la manière dont les faits doivent entrer dans l'évaluation normative, et détermine les limites de la subsomption ; doit être convenablement distinguée de l'interprétation téléologique (selon le but de la norme) au sein de l'interprétation juridique." }, "ja": { "name": "normative-meaning-argumentation", "summary": "法規範の目的と価値指向、そして事実がどのように規範的評価に入るべきかを分析し、当てはめ（涵摂）の限界を判断します。法律解釈における規範目的解釈とは適切に区別する必要があります。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:administrative-value-judgment", "slug": "skill-administrative-value-judgment", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Assists legal staff of mainland Chinese administrative agencies in making value judgments and balancing interests under the fundamental principles of administrative law, based on a description of an administrative case, and in forming an indicative discretionary conclusion." }, "human": { "zh-CN": { "name": "administrative-value-judgment", "summary": "辅助中国大陆行政机关法律工作人员，根据一段行政案情，按行政法基本原则进行价值判断、利益衡量，并形成倾向性裁量结论。" }, "en": { "name": "administrative-value-judgment", "summary": "Assists legal staff of mainland Chinese administrative agencies in making value judgments and balancing interests under the fundamental principles of administrative law, based on a description of an administrative case, and in forming an indicative discretionary conclusion." }, "es": { "name": "administrative-value-judgment", "summary": "Asiste al personal jurídico de los organismos administrativos de China continental a realizar juicios de valor y ponderar intereses conforme a los principios fundamentales del derecho administrativo, a partir de la descripción de un caso administrativo, y a formar una conclusión indicativa sobre el ejercicio de la discrecionalidad." }, "fr": { "name": "administrative-value-judgment", "summary": "Aide le personnel juridique des autorités administratives de Chine continentale à porter des jugements de valeur et à pondérer les intérêts selon les principes fondamentaux du droit administratif, à partir de l'exposé d'un cas administratif, et à former une conclusion d'appréciation discrétionnaire indicative." }, "ja": { "name": "administrative-value-judgment", "summary": "中国大陸の行政機関の法務担当者を支援し、行政事案の記述に基づいて行政法の基本原則に沿った価値判断と利益衡量を行い、方向性のある裁量結論を形成します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:analogical-reasoning", "slug": "skill-analogical-reasoning", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when the AI agent needs to handle the following situations: 1. The current case/issue has no direct, explicit provision in existing law (a legal gap); 2. Existing precedents, similar cases, or related legal norms must be consulted to derive a solution for the current issue; 3. The user expl" }, "human": { "zh-CN": { "name": "analogical-reasoning", "summary": "当AI智能体需要处理以下情形时触发本技能： 1. 当前案件/问题在现行法律中没有直接、明确的规定（法律漏洞）； 2. 需要参考已有判例、相似案件或相近法律规范来推导当前问题的处理方案； 3. 用户明确要求\"参照\"\"比照\"\"类推适用\"某一规范或案例； 4. 在法律论证中需要以\"相似案件相似处理\"（Like cases should be treated alike）原则支撑结论； 5. 需要评估两个或多个案件/规范之间的相似性与差异性，以判断能否进行类比； 6. 在合同解释、侵权责任认定、行政处罚裁量等场景中，需要借助类似情形的处理经验形成论证。 本技能的核心是：识别相似性基础（tertium comparationis），论证类比的正当性，并在此基础上得出合理的法律结论。" }, "en": { "name": "analogical-reasoning", "summary": "Triggers when the AI agent needs to handle the following situations: 1. The current case/issue has no direct, explicit provision in existing law (a legal gap); 2. Existing precedents, similar cases, or related legal norms must be consulted to derive a solution for the current issue; 3. The user explicitly requests \"referring to\", \"applying by correspondence\", or \"analogical application of\" a norm or case; 4. A legal argument needs the principle that \"like cases should be treated alike\" to support its conclusion; 5. The similarity and differences between two or more cases/norms must be assessed to judge whether an analogy can be drawn; 6. In contract interpretation, tort liability determination, administrative penalty discretion, and similar scenarios, experience from analogous situations is needed to build the argument. The core of this skill is to identify the basis of similarity (tertium comparationis), justify the analogy, and on that basis reach a sound legal conclusion." }, "es": { "name": "analogical-reasoning", "summary": "Se activa cuando el agente de IA necesita gestionar las siguientes situaciones: 1. El caso/cuestión actual carece de disposición directa y explícita en la ley vigente (laguna legal); 2. Deben consultarse precedentes, casos similares o normas jurídicas afines para derivar una solución al problema actual; 3. El usuario solicita expresamente \"referirse a\", \"aplicar por correspondencia\" o \"aplicar por analogía\" una norma o caso; 4. La argumentación jurídica necesita el principio de que \"los casos semejantes deben tratarse igual\" para sostener la conclusión; 5. Deben evaluarse las semejanzas y diferencias entre dos o más casos/normas para juzgar si procede la analogía; 6. En interpretación de contratos, determinación de la responsabilidad extracontractual, discrecionalidad sancionadora administrativa y otros escenarios, se necesita la experiencia de situaciones análogas para construir el argumento. El núcleo de esta habilidad es identificar el fundamento de semejanza (tertium comparationis), justificar la analogía y, sobre esa base, llegar a una conclusión jurídica razonable." }, "fr": { "name": "analogical-reasoning", "summary": "Se déclenche lorsque l'agent IA doit traiter les situations suivantes : 1. Le cas/problème actuel n'a pas de disposition directe et explicite dans le droit en vigueur (lacune juridique) ; 2. Des précédents, des affaires similaires ou des normes juridiques proches doivent être consultés pour déduire une solution au problème actuel ; 3. L'utilisateur demande explicitement de « se référer à », « appliquer par correspondance » ou « appliquer par analogie » une norme ou une affaire ; 4. L'argumentation juridique a besoin du principe « les cas semblables doivent être traités de la même manière » (like cases should be treated alike) pour étayer la conclusion ; 5. Les similitudes et différences entre deux cas/normes ou plus doivent être évaluées pour juger si l'analogie est possible ; 6. Dans l'interprétation contractuelle, la détermination de la responsabilité délictuelle, l'appréciation des sanctions administratives, etc., l'expérience de situations analogues est nécessaire pour construire l'argumentation. Le cœur de cette compétence : identifier le fondement de la similitude (tertium comparationis), justifier l'analogie et, sur cette base, parvenir à une conclusion juridique raisonnable." }, "ja": { "name": "analogical-reasoning", "summary": "AIエージェントが以下の状況を扱う必要があるときに起動します：1. 現在の事件・問題について現行法に直接かつ明確な規定がない場合（法の欠落）；2. 既存の判例、類似事件または近接する法規範を参照して解決策を導く必要がある場合；3. ユーザーが「〜を参照」「〜に準拠」「類推適用」を明示的に求める場合；4. 法的論証において「類似事案は類似に処理する」（Like cases should be treated alike）の原則で結論を支える必要がある場合；5. 二つ以上の事件・規範間の類似性と相違性を評価し、類推の可否を判断する必要がある場合；6. 契約解釈、不法行為責任の認定、行政処分の裁量などの場面で、類似事案の処理経験を論証に活かす必要がある場合。本スキルの核心は、類似性の基礎（tertium comparationis）を特定し、類推の正当性を論証し、その上で妥当な法的結論を導くことです。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:argument-chain-construction", "slug": "skill-argument-chain-construction", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when the AI agent needs to organize legal reasoning results into a complete, self-consistent, persuasive argument structure. Typical trigger scenarios include but are not limited to: drafting the argumentation sections of legal documents (legal opinions, statements of representation, defens" }, "human": { "zh-CN": { "name": "argument-chain-construction", "summary": "当AI智能体需要将法律推理结果组织为完整、自洽、有说服力的论证结构时触发此技能。 典型触发场景包括但不限于： - 撰写法律意见书、代理词、辩护词、答辩状等法律文书中的论证部分 - 回答用户提出的复杂法律问题，需要多步骤推理并呈现结论 - 对争议焦点进行系统性分析，需要构建从事实到法律到结论的完整推理路径 - 反驳对方观点时，需要构建替代性论证链条 - 对多个法律问题进行综合分析，需要确保各论证之间逻辑一致、互不矛盾 - 在法律研究备忘录中呈现分析结论 触发信号关键词：论证、分析、推理、为什么、证明、理由、论述、阐述观点、法律分析" }, "en": { "name": "argument-chain-construction", "summary": "Triggers when the AI agent needs to organize legal reasoning results into a complete, self-consistent, persuasive argument structure. Typical trigger scenarios include but are not limited to: drafting the argumentation sections of legal documents (legal opinions, statements of representation, defense statements, statements of defense); answering complex legal questions that require multi-step reasoning and a presented conclusion; systematically analyzing disputed points, requiring a complete reasoning path from facts to law to conclusion; constructing an alternative argument chain when rebutting the other side's view; comprehensively analyzing multiple legal issues while ensuring logical consistency and non-contradiction across arguments; presenting analysis conclusions in legal research memoranda. Trigger keywords: argumentation, analysis, reasoning, why, proof, grounds, exposition, articulating a viewpoint, legal analysis." }, "es": { "name": "argument-chain-construction", "summary": "Se activa cuando el agente de IA necesita organizar los resultados del razonamiento jurídico en una estructura argumentativa completa, coherente y persuasiva. Escenarios típicos de activación incluyen, entre otros: redactar las secciones argumentativas de documentos jurídicos (dictámenes, alegatos de representación, alegatos de defensa, contestaciones); responder a preguntas jurídicas complejas que requieren razonamiento en varios pasos y presentación de conclusiones; analizar sistemáticamente los puntos controvertidos, construyendo una ruta completa de razonamiento de los hechos al derecho y a la conclusión; construir una cadena argumentativa alternativa al refutar el punto de vista contrario; analizar de forma integral varios problemas jurídicos asegurando la coherencia lógica y la ausencia de contradicciones entre argumentos; presentar conclusiones de análisis en memorandos de investigación jurídica. Palabras clave de activación: argumentación, análisis, razonamiento, por qué, prueba, fundamentos, exposición, articular un punto de vista, análisis jurídico." }, "fr": { "name": "argument-chain-construction", "summary": "Se déclenche lorsque l'agent IA doit organiser les résultats du raisonnement juridique en une structure argumentative complète, cohérente et convaincante. Scénarios types de déclenchement, sans s'y limiter : rédiger les parties argumentatives des documents juridiques (avis juridiques, conclusions de représentation, plaidoiries de défense, conclusions en défense) ; répondre à des questions juridiques complexes exigeant un raisonnement en plusieurs étapes et une conclusion exposée ; analyser systématiquement les points litigieux en construisant un chemin de raisonnement complet des faits au droit puis à la conclusion ; bâtir une chaîne argumentative alternative pour réfuter la thèse adverse ; analyser de façon intégrée plusieurs questions juridiques en assurant la cohérence logique entre les arguments ; présenter des conclusions d'analyse dans des mémos de recherche juridique. Mots-clés de déclenchement : argumentation, analyse, raisonnement, pourquoi, preuve, motifs, exposé, articuler un point de vue, analyse juridique." }, "ja": { "name": "argument-chain-construction", "summary": "AIエージェントが法的推論の結果を、完全で自己整合的かつ説得力のある論証構造に組織する必要があるときに起動します。典型的な起動場面（これらに限りません）：法律意見書、代理人意見書、弁護意見書、答弁書などの法律文書の論証部分の執筆；複数ステップの推論と結論の提示を要する複雑な法律問題への回答；争点の体系的な分析と、事実から法律、結論に至る完全な推論経路の構築；相手方の見解を反駁する際の代替的な論証連鎖の構築；複数の法律問題の統合分析と、論証間の論理的整合性・無矛盾性の確保；法律リサーチメモでの分析結論の提示。起動信号キーワード：論証、分析、推論、なぜ、証明、理由、論述、見解の展開、法律分析。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:argument-strength-evaluation", "slug": "skill-argument-strength-evaluation", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers after the AI agent completes a piece of legal reasoning or argumentation and needs to self-assess the reasoning conclusion, gauge the overall strength and confidence of the argument, and identify and flag weak links in the reasoning chain. Trigger conditions include but are not limited to: " }, "human": { "zh-CN": { "name": "argument-strength-evaluation", "summary": "当AI智能体完成一段法律推理或论证后，需要对该推理结论进行自我评估，判断论证的整体强度与置信度，识别并标注推理链条中的薄弱环节。 触发条件包括但不限于： - 完成法律分析后需要给出结论可靠性评级 - 用户明确要求评估某一论证的说服力或可靠性 - 存在多个竞争性法律解释需要比较论证强度 - 推理过程中涉及不确定因素（事实不明、法律空白、争议观点）需要标注 - 为决策者提供风险提示，说明结论的局限性 - 对己方或对方的法律论证进行攻防强度评估 - 在法律意见书、备忘录等文书中需要附加置信度说明" }, "en": { "name": "argument-strength-evaluation", "summary": "Triggers after the AI agent completes a piece of legal reasoning or argumentation and needs to self-assess the reasoning conclusion, gauge the overall strength and confidence of the argument, and identify and flag weak links in the reasoning chain. Trigger conditions include but are not limited to: assigning a reliability rating to a conclusion after legal analysis; the user explicitly asking to evaluate the persuasiveness or reliability of an argument; multiple competing legal interpretations whose argument strength must be compared; uncertainties in the reasoning (unclear facts, legal gaps, disputed views) that need flagging; providing decision-makers with risk warnings about a conclusion's limitations; assessing the offensive/defensive strength of one's own or the opponent's legal arguments; attaching a confidence statement in documents such as legal opinions and memoranda." }, "es": { "name": "argument-strength-evaluation", "summary": "Se activa cuando el agente de IA completa un razonamiento o argumento jurídico y necesita autoevaluar la conclusión, calibrar la fuerza global y el nivel de confianza del argumento, e identificar y señalar los eslabones débiles de la cadena de razonamiento. Condiciones de activación incluyen, entre otras: asignar una valoración de fiabilidad a la conclusión tras el análisis jurídico; que el usuario pida explícitamente evaluar la persuasividad o fiabilidad de un argumento; la existencia de varias interpretaciones jurídicas rivales cuyas fuerzas argumentativas deban compararse; incertidumbres en el razonamiento (hechos poco claros, vacíos legales, opiniones en disputa) que deban señalarse; advertir a los decisores sobre las limitaciones de la conclusión; evaluar la fuerza ofensiva/defensiva de los argumentos propios o de la contraparte; añadir una nota de confianza en documentos como dictámenes y memorandos jurídicos." }, "fr": { "name": "argument-strength-evaluation", "summary": "Se déclenche lorsque l'agent IA a achevé un raisonnement ou une argumentation juridique et doit s'auto-évaluer : mesurer la force globale et le degré de confiance de l'argument, repérer et signaler les maillons faibles de la chaîne de raisonnement. Conditions de déclenchement, sans s'y limiter : attribuer un niveau de fiabilité à la conclusion après l'analyse juridique ; l'utilisateur demande explicitement d'évaluer la force de persuasion ou la fiabilité d'un argument ; plusieurs interprétations juridiques concurrentes dont il faut comparer la force argumentative ; incertitudes dans le raisonnement (faits obscurs, lacunes juridiques, opinions contestées) à signaler ; avertir les décideurs des limites de la conclusion ; évaluer la force offensive/défensive des arguments de son camp ou de la partie adverse ; joindre une note de confiance aux avis juridiques, mémos et autres documents." }, "ja": { "name": "argument-strength-evaluation", "summary": "AIエージェントが法的推論や論証を完了した後、その推論結論を自己評価し、論証全体の強度と信頼度を判断し、推論連鎖の中の弱点を特定・標示する必要があるときに起動します。起動条件（これらに限りません）：法律分析後に結論の信頼性評価を付ける；ユーザーが特定の論証の説得力や信頼性の評価を明示的に求める；複数の競合する法律解釈間で論証強度を比較する；推論過程の不確定要素（事実不明瞭、法の欠落、争いのある見解）を標示する；意思決定者に結論の限界をリスク通知として示す；自陣または相手方の法的論証の攻防強度を評価する；法律意見書やメモなどの文書に信頼度の説明を付す。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:billing-and-litigation-budget", "slug": "skill-billing-and-litigation-budget", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Use this skill when the user needs to tally and manage attorney work hours, expert fees, and investigation costs; control litigation costs; or issue timesheets or fee statements to clients. Typical trigger scenarios include but are not limited to: mentioning keywords such as \"work hours\", \"expert fe" }, "human": { "zh-CN": { "name": "billing-and-litigation-budget", "summary": "当用户需要统计、管理律师工时、专家费用、调查费用，控制诉讼成本、向客户出具工时单或费用单时使用此技能。 典型触发场景包括但不限于： - 提到\"工作小时\"\"专家费用\"\"调查费用\"\"办案成本\"\"工时单\"\"费用单\"\"预算\"\"成本控制\"等关键词 - 询问某一律师的工作小时、某一案件的办案费用或诉讼成本时 - 用户要求交付律师工作时长总结或办案工时单、办案费用单，用于交付客户或申请报销时 - 需要对某一案件或常法服务进行预算编制、预算执行监控或预算偏差分析时 - 需要评估诉讼成本与预期收益，进行诉讼经济性分析时" }, "en": { "name": "billing-and-litigation-budget", "summary": "Use this skill when the user needs to tally and manage attorney work hours, expert fees, and investigation costs; control litigation costs; or issue timesheets or fee statements to clients. Typical trigger scenarios include but are not limited to: mentioning keywords such as \"work hours\", \"expert fees\", \"investigation costs\", \"case-handling costs\", \"timesheet\", \"fee statement\", \"budget\", or \"cost control\"; asking about a given attorney's work hours or a given case's handling costs or litigation costs; requesting a summary of attorney time or case timesheets/fee statements for delivery to a client or for reimbursement claims; budgeting, budget-execution monitoring, or budget variance analysis for a case or for ongoing general legal services; and evaluating litigation costs against expected returns for a litigation economics analysis." }, "es": { "name": "billing-and-litigation-budget", "summary": "Use esta habilidad cuando el usuario necesite contabilizar y gestionar horas de trabajo de abogados, honorarios de peritos y gastos de investigación; controlar costos de litigio; o emitir partes de horas o estados de honorarios a clientes. Escenarios típicos de activación incluyen, entre otros: mencionar palabras clave como \"horas de trabajo\", \"honorarios de peritos\", \"gastos de investigación\", \"costos de tramitación\", \"parte de horas\", \"estado de honorarios\", \"presupuesto\" o \"control de costos\"; preguntar por las horas de trabajo de un abogado o por los costos de tramitación o de litigio de un caso; solicitar un resumen del tiempo de trabajo de los abogados o partes de horas y estados de honorarios para entregar al cliente o pedir reembolso; elaborar presupuestos, monitorear su ejecución o analizar desviaciones para un caso o para servicios jurídicos permanentes; y evaluar costos de litigio frente a rendimientos esperados para un análisis de economía del litigio." }, "fr": { "name": "billing-and-litigation-budget", "summary": "Utilisez cette compétence lorsque l'utilisateur doit recenser et gérer les heures de travail des avocats, les honoraires d'experts et les frais d'enquête ; maîtriser les coûts du contentieux ; ou établir des feuilles d'heures ou des relevés d'honoraires à l'attention des clients. Scénarios types de déclenchement, sans s'y limiter : mention de mots-clés tels que « heures de travail », « honoraires d'experts », « frais d'enquête », « coûts de traitement du dossier », « feuille d'heures », « relevé d'honoraires », « budget », « maîtrise des coûts » ; question sur les heures d'un avocat ou sur les coûts de traitement/contentieux d'un dossier ; demande d'un récapitulatif du temps des avocats ou de feuilles d'heures et relevés d'honoraires à remettre au client ou pour une demande de remboursement ; élaboration d'un budget, suivi de son exécution ou analyse des écarts pour un dossier ou des services juridiques récurrents ; évaluation des coûts du contentieux par rapport aux gains attendus pour une analyse de l'économie du litige." }, "ja": { "name": "billing-and-litigation-budget", "summary": "ユーザーが弁護士の作業時間、専門家費用、調査費用を集計・管理したり、訴訟コストを管理したり、タイムシート（工数表）や費用明細をクライアントに発行したりする必要があるときに使用します。典型的な起動場面（これらに限りません）：「作業時間」「専門家費用」「調査費用」「事件処理コスト」「タイムシート」「費用明細」「予算」「コスト管理」などのキーワードへの言及；特定の弁護士の作業時間、特定の事件の処理費用や訴訟コストへの照会；クライアントへの納品や経費精算のために、弁護士作業時間の総括や事件のタイムシート・費用明細の提出を求められたとき；特定の事件または常設法務サービスの予算編成、予算執行のモニタリング、予算差異分析が必要なとき；訴訟コストと期待収益を評価し、訴訟経済性分析を行う必要があるとき。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:case-lifecycle-planning", "slug": "skill-case-lifecycle-planning", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Case full-lifecycle planning skill. Triggers when the user needs to plan a case preparation timeline, create or ask about case preparation milestones, or generate a litigation roadmap and a key-dates overview table. Typical trigger keywords: \"case preparation\", \"case preparation planning\", \"case pre" }, "human": { "zh-CN": { "name": "case-lifecycle-planning", "summary": "案件全周期规划技能。当用户需要规划案件准备时间线、创建或询问案件准备时间节点、生成案件诉讼路线图和关键时间一览表时触发。典型触发关键词：\"案件准备\"\"案件准备规划\"\"案件准备时间线\"\"规划案件准备节奏\"\"排开庭\"。不处理案件向律师的分配，也不处理与案件管理无关的一般性编程任务。" }, "en": { "name": "case-lifecycle-planning", "summary": "Case full-lifecycle planning skill. Triggers when the user needs to plan a case preparation timeline, create or ask about case preparation milestones, or generate a litigation roadmap and a key-dates overview table. Typical trigger keywords: \"case preparation\", \"case preparation planning\", \"case preparation timeline\", \"plan the case preparation pace\", \"schedule the hearing\". Does not handle assigning cases to attorneys, nor general programming tasks unrelated to case management." }, "es": { "name": "case-lifecycle-planning", "summary": "Habilidad de planificación del ciclo de vida completo del caso. Se activa cuando el usuario necesita planificar la línea de tiempo de preparación del caso, crear o consultar hitos de preparación, o generar una hoja de ruta del litigio y un cuadro de fechas clave. Palabras clave típicas de activación: \"preparación del caso\", \"planificación de la preparación del caso\", \"línea de tiempo de preparación del caso\", \"planificar el ritmo de preparación\", \"agendar la audiencia\". No gestiona la asignación de casos a abogados ni tareas de programación general ajenas a la gestión de casos." }, "fr": { "name": "case-lifecycle-planning", "summary": "Compétence de planification du cycle de vie complet du dossier. Se déclenche lorsque l'utilisateur doit planifier la chronologie de préparation du dossier, créer ou consulter des jalons de préparation, ou produire une feuille de route contentieuse et un tableau des dates clés. Mots-clés typiques de déclenchement : « préparation du dossier », « planification de la préparation du dossier », « chronologie de préparation du dossier », « planifier le rythme de préparation », « fixer l'audience ». Ne gère ni l'attribution des dossiers aux avocats, ni les tâches de programmation générale étrangères à la gestion des dossiers." }, "ja": { "name": "case-lifecycle-planning", "summary": "事件の全ライフサイクル計画スキルです。ユーザーが事件準備のタイムラインを計画したり、準備のマイルストーンを作成・照会したり、訴訟ロードマップと主要日程の一覧表を生成したりする必要があるときに起動します。典型的な起動キーワード：「事件準備」「事件準備の計画」「事件準備のタイムライン」「準備ペースの計画」「公判期日の日程調整」。事件の弁護士への割り当てや、事件管理に関係しない一般的なプログラミング作業は扱いません。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:case-retrieval", "slug": "skill-case-retrieval", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when the user needs to find similar cases, related judgments, or adjudication rules relevant to the current legal issue. Typical trigger scenarios include: the user explicitly asks to find similar cases or precedents; past rulings are needed to support a legal argument; predicting a case ou" }, "human": { "zh-CN": { "name": "case-retrieval", "summary": "当用户需要查找与当前法律问题相关的类似案例、相关判决、裁判规则时触发此技能。典型触发场景包括：用户明确要求查找类似案例或判例；需要通过既往裁判支撑法律论点；进行案件预判需要参考同类判决；撰写法律文书需要引用权威案例；分析某一法律问题的司法实践趋势；对比不同法院、不同时期对同一问题的裁判立场。关键词包括但不限于：类似案例、相关判决、判例检索、裁判规则、同类案件、指导性案例、典型案例、司法观点、裁判要旨等。" }, "en": { "name": "case-retrieval", "summary": "Triggers when the user needs to find similar cases, related judgments, or adjudication rules relevant to the current legal issue. Typical trigger scenarios include: the user explicitly asks to find similar cases or precedents; past rulings are needed to support a legal argument; predicting a case outcome requires reference to judgments of like cases; drafting legal documents requires citing authoritative cases; analyzing the trends of judicial practice on a legal issue; comparing the adjudication positions of different courts or periods on the same issue. Keywords include but are not limited to: similar cases, related judgments, precedent search, adjudication rules, like cases, guiding cases, model cases, judicial views, gist of rulings, and the like." }, "es": { "name": "case-retrieval", "summary": "Se activa cuando el usuario necesita encontrar casos similares, sentencias relacionadas o criterios de decisión pertinentes al problema jurídico actual. Escenarios típicos de activación: el usuario pide explícitamente buscar casos o precedentes similares; se necesitan resoluciones previas para respaldar un argumento jurídico; anticipar el desenlace de un caso requiere referirse a fallos de casos análogos; redactar documentos jurídicos exige citar casos que hacen autoridad; analizar las tendencias de la práctica judicial sobre un tema; comparar las posturas de distintos tribunales o épocas sobre el mismo asunto. Palabras clave incluyen, entre otras: casos similares, sentencias relacionadas, búsqueda de precedentes, criterios de decisión, casos análogos, casos guía, casos paradigmáticos, criterios judiciales, doctrina esencial de los fallos, etc." }, "fr": { "name": "case-retrieval", "summary": "Se déclenche lorsque l'utilisateur doit trouver des affaires similaires, des décisions connexes ou des règles de jugement liées au problème juridique en cours. Scénarios types de déclenchement : l'utilisateur demande explicitement de trouver des affaires ou précédents similaires ; des décisions passées sont nécessaires pour étayer un argument juridique ; anticiper l'issue d'un dossier exige de se référer à des décisions d'affaires analogues ; rédiger des documents juridiques impose de citer des affaires faisant autorité ; analyser les tendances de la pratique judiciaire sur une question ; comparer les positions de différents tribunaux ou périodes sur un même problème. Mots-clés, sans s'y limiter : affaires similaires, décisions connexes, recherche de jurisprudence, règles de jugement, affaires analogues, affaires directrices, affaires types, positions judiciaires, motifs essentiels des décisions, etc." }, "ja": { "name": "case-retrieval", "summary": "ユーザーが現在の法律問題に関連する類似事例、関連判決、裁判規則を調べる必要があるときに起動します。典型的な起動場面：類似事例や判例の検索を明示的に求める場合；過去の裁判で法的論点を支える必要がある場合；事件の見通しを予測するのに同種判決を参照する必要がある場合；法律文書の起草で権威ある事例を引用する必要がある場合；ある法律問題の司法実務の動向を分析する場合；同じ問題に対する裁判所別・時期別の裁判立場を比較する場合。キーワードには（これらに限りません）：類似事例、関連判決、判例検索、裁判規則、同種事件、指導案例、典型事件、司法見解、裁判要旨など。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:conflict-resolution", "slug": "skill-conflict-resolution", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when the AI agent encounters the following situations during legal reasoning: (1) multiple legal provisions can apply to the same facts (concurrence of provisions); (2) contradictions or inconsistencies among different pieces of evidence (evidence conflict); (3) multiple disputed issues req" }, "human": { "zh-CN": { "name": "conflict-resolution", "summary": "当AI智能体在法律推理过程中遇到以下情形时触发本技能：(1)多个法律条文均可适用于同一事实（法条竞合）；(2)不同证据之间存在矛盾或不一致（证据矛盾）；(3)案件存在多个争议焦点需要确定审理或分析顺序（争点优先级排序）；(4)不同法律渊源（法律、法规、司法解释、地方性法规等）对同一问题作出不同规定（法源冲突）；(5)当事人主张与法律适用之间存在张力需要协调。本技能是法律推理的核心枢纽能力，几乎所有复杂法律分析都会涉及某种形式的冲突解决。" }, "en": { "name": "conflict-resolution", "summary": "Triggers when the AI agent encounters the following situations during legal reasoning: (1) multiple legal provisions can apply to the same facts (concurrence of provisions); (2) contradictions or inconsistencies among different pieces of evidence (evidence conflict); (3) multiple disputed issues requiring an order of trial or analysis (prioritization of issues); (4) different sources of law (statutes, regulations, judicial interpretations, local regulations, etc.) providing differently on the same issue (conflict of legal sources); (5) tension between a party's claims and the applicable law that needs reconciling. This skill is the core hub capability of legal reasoning; almost all complex legal analysis involves some form of conflict resolution." }, "es": { "name": "conflict-resolution", "summary": "Se activa cuando el agente de IA encuentra las siguientes situaciones durante el razonamiento jurídico: (1) varios preceptos legales pueden aplicarse a los mismos hechos (concurrencia de preceptos); (2) contradicciones o inconsistencias entre distintas pruebas (contradicción probatoria); (3) múltiples puntos controvertidos que exigen un orden de examen o análisis (priorización de cuestiones); (4) distintas fuentes del derecho (leyes, reglamentos, interpretaciones judiciales, normas locales, etc.) que disponen de forma diferente sobre un mismo asunto (conflicto de fuentes); (5) tensión entre las pretensiones de las partes y el derecho aplicable que requiere conciliación. Esta habilidad es la capacidad nodal central del razonamiento jurídico; casi todo análisis jurídico complejo implica alguna forma de resolución de conflictos." }, "fr": { "name": "conflict-resolution", "summary": "Se déclenche lorsque l'agent IA rencontre les situations suivantes en raisonnement juridique : (1) plusieurs dispositions légales peuvent s'appliquer aux mêmes faits (concours de dispositions) ; (2) contradictions ou incohérences entre différentes preuves (contradiction probatoire) ; (3) plusieurs points litigieux exigeant un ordre de traitement ou d'analyse (hiérarchisation des questions) ; (4) différentes sources de droit (lois, règlements, interprétations judiciaires, règlements locaux, etc.) qui disposent différemment sur un même problème (conflit de sources) ; (5) tension entre les prétentions des parties et le droit applicable qu'il faut concilier. Cette compétence est la capacité pivot du raisonnement juridique : presque toute analyse juridique complexe implique une forme de résolution de conflits." }, "ja": { "name": "conflict-resolution", "summary": "AIエージェントが法律推論の過程で以下の状況に直面したときに起動します：(1) 複数の法律条文が同一事実に適用可能な場合（法条競合）；(2) 異なる証拠間に矛盾や不一致がある場合（証拠矛盾）；(3) 複数の争点があり審理・分析の順序を決める必要がある場合（争点の優先順位付け）；(4) 法源（法律、法規、司法解釈、地方性法規など）ごとに同一問題への規定が異なる場合（法源の衝突）；(5) 当事者の主張と法律適用の間に調整すべき緊張関係がある場合。本スキルは法律推論の中核となるハブ能力であり、ほぼすべての複雑な法律分析に何らかの形の衝突解決が関わります。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:counterfactual-reasoning", "slug": "skill-counterfactual-reasoning", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when one needs to assess \"how the legal outcome would differ if a given fact/action/condition did not exist or changed\". Typical trigger scenarios include: causation determination (would the harm still occur absent the act), apportioning fault and liability shares, defining the scope of dam" }, "human": { "zh-CN": { "name": "counterfactual-reasoning", "summary": "当需要评估'如果某一事实/行为/条件不存在或发生改变，法律结果将如何不同'时触发此技能。典型触发场景包括：因果关系认定（若无该行为，损害是否仍会发生）、过错与责任比例划分、损害赔偿范围确定、合同违约损失计算、行政行为合法性假设检验、刑事因果关系中断判断、立法影响评估等。核心方法是构建一个与现实世界仅在目标变量上不同的'反事实世界'，比较两个世界的结果差异，从而得出目标变量对结果的贡献程度。" }, "en": { "name": "counterfactual-reasoning", "summary": "Triggers when one needs to assess \"how the legal outcome would differ if a given fact/action/condition did not exist or changed\". Typical trigger scenarios include: causation determination (would the harm still occur absent the act), apportioning fault and liability shares, defining the scope of damages, calculating contract breach losses, hypothesis-testing the legality of administrative acts, judging interruptions of criminal causation, assessing legislative impact, and the like. The core method is to construct a \"counterfactual world\" that differs from the real world only in the target variable, compare the outcome differences between the two worlds, and thereby derive the target variable's degree of contribution to the outcome." }, "es": { "name": "counterfactual-reasoning", "summary": "Se activa cuando hay que evaluar \"cómo cambiaría el resultado jurídico si un determinado hecho/actuación/condición no existiera o cambiara\". Escenarios típicos de activación: determinación de la causalidad (¿se produciría igualmente el daño sin esa actuación?), reparto de la culpa y de las proporciones de responsabilidad, delimitación del alcance de la indemnización, cálculo de pérdidas por incumplimiento contractual, prueba de hipótesis sobre la legalidad de actos administrativos, juicio sobre la interrupción de la causalidad penal, evaluación del impacto legislativo, etc. El método central consiste en construir un \"mundo contrafáctico\" que difiera del mundo real solo en la variable objetivo, comparar las diferencias de resultados entre ambos mundos y así obtener el grado de contribución de la variable objetivo al resultado." }, "fr": { "name": "counterfactual-reasoning", "summary": "Se déclenche lorsqu'il faut évaluer « comment le résultat juridique changerait si un fait/action/condition donné n'existait pas ou changeait ». Scénarios types de déclenchement : établissement de la causalité (le dommage se produirait-il malgré l'absence de l'acte ?), répartition des fautes et des parts de responsabilité, délimitation du périmètre d'indemnisation, calcul des pertes liées à une inexécution contractuelle, test d'hypothèse sur la légalité d'un acte administratif, appréciation de la rupture de la causalité pénale, évaluation de l'impact d'une législation, etc. La méthode centrale consiste à construire un « monde contrefactuel » ne différant du monde réel que par la variable cible, à comparer les écarts de résultats entre les deux mondes et à en déduire le degré de contribution de la variable cible au résultat." }, "ja": { "name": "counterfactual-reasoning", "summary": "「ある事実・行為・条件が存在しなかったり変化したりした場合、法的結果がどう変わるか」を評価する必要があるときに起動します。典型的な起動場面：因果関係の認定（その行為がなければ損害は発生したか）、過失と責任割合の配分、損害賠償範囲の確定、契約違反の損失計算、行政行為の合法性の仮説検証、刑事因果関係の中断判断、立法影響評価など。核心メソッドは、現実世界と目標変数のみが異なる「反実仮想世界」を構築し、二つの世界の結果の差を比較することで、目標変数が結果に及ぼす寄与の程度を導くことです。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:deductive-reasoning", "slug": "skill-deductive-reasoning", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Performs rigorous legal deductive reasoning based on formal logic. Use this skill when the user asks to \"analyze case logic\", \"verify legal reasoning\", \"extract syllogisms\", \"map the relationship between case facts and legal provisions\", or \"judge the validity of a legal argument\". Provides the user" }, "human": { "zh-CN": { "name": "deductive-reasoning", "summary": "基于形式逻辑执行严密的法律演绎推理。当用户要求\"分析案件逻辑\"、\"验证法律推理\"、\"提取三段论\"、\"梳理案件事实与法条的关系\"或\"判断法律论证的有效性\"时使用本技能。 为用户提供基于形式逻辑的法律推理框架，将非结构化的法律规范与案件事实转化为有效、可检验、可复现的演绎论证链条（P-F-C）。" }, "en": { "name": "deductive-reasoning", "summary": "Performs rigorous legal deductive reasoning based on formal logic. Use this skill when the user asks to \"analyze case logic\", \"verify legal reasoning\", \"extract syllogisms\", \"map the relationship between case facts and legal provisions\", or \"judge the validity of a legal argument\". Provides the user with a formal-logic-based legal reasoning framework, converting unstructured legal norms and case facts into valid, testable, reproducible deductive argument chains (P-F-C)." }, "es": { "name": "deductive-reasoning", "summary": "Ejecuta un razonamiento jurídico deductivo riguroso basado en lógica formal. Use esta habilidad cuando el usuario pida \"analizar la lógica del caso\", \"verificar el razonamiento jurídico\", \"extraer silogismos\", \"ordenar la relación entre los hechos del caso y los preceptos\" o \"juzgar la validez de un argumento jurídico\". Ofrece al usuario un marco de razonamiento jurídico basado en lógica formal, convirtiendo normas jurídicas y hechos del caso no estructurados en cadenas argumentativas deductivas válidas, verificables y reproducibles (P-F-C)." }, "fr": { "name": "deductive-reasoning", "summary": "Exécute un raisonnement juridique déductif rigoureux fondé sur la logique formelle. Utilisez cette compétence lorsque l'utilisateur demande d'« analyser la logique du dossier », de « vérifier le raisonnement juridique », d'« extraire des syllogismes », de « mettre en ordre la relation entre les faits du dossier et les dispositions » ou de « juger la validité d'un argument juridique ». Fournit à l'utilisateur un cadre de raisonnement juridique fondé sur la logique formelle, convertissant normes juridiques et faits de dossier non structurés en chaînes argumentatives déductives valides, vérifiables et reproductibles (P-F-C)." }, "ja": { "name": "deductive-reasoning", "summary": "形式論理に基づく厳密な法的演繹推論を実行します。ユーザーが「事件のロジックを分析して」「法律推論を検証して」「三段論法を抽出して」「事件事実と条文の関係を整理して」「法律論証の有効性を判断して」と求めたときに本スキルを使用します。形式論理に基づく法律推論の枠組みを提供し、非構造化された法律規範と事件事実を、有効で検証可能・再現可能な演繹論証連鎖（P-F-C）へと変換します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:dispute-and-performance-risk", "slug": "skill-dispute-and-performance-risk", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when the user submits contract texts, transaction arrangements, commercial agreements, or descriptions of legal relations and needs to assess potential legal disputes or breach risks within them. Typical trigger scenarios include but are not limited to: the user asks to review a contract an" }, "human": { "zh-CN": { "name": "dispute-and-performance-risk", "summary": "当用户提交合同文本、交易安排、商业协议或法律关系描述，需要评估其中可能产生法律纠纷或违约风险时触发此技能。 典型触发场景包括但不限于： - 用户要求审查合同并指出潜在风险 - 用户询问某项交易安排是否存在履约障碍 - 用户描述合同执行中的困难，需要判断是否构成违约 - 用户要求对合同条款进行风险评级 - 用户提供两方或多方之间的协议，要求识别可能的争议焦点 - 用户描述已发生的履约偏差，需要评估法律后果 - 合同审查、尽职调查、交易谈判等场景中需要前瞻性风险识别 此技能聚焦于\"是否会产生法律纠纷\"和\"是否存在违约风险\"两大核心问题， 输出结构化的风险清单与应对建议。" }, "en": { "name": "dispute-and-performance-risk", "summary": "Triggers when the user submits contract texts, transaction arrangements, commercial agreements, or descriptions of legal relations and needs to assess potential legal disputes or breach risks within them. Typical trigger scenarios include but are not limited to: the user asks to review a contract and flag potential risks; asks whether a transaction arrangement faces performance obstacles; describes difficulties in contract execution and needs to determine whether it constitutes breach; asks for risk ratings of contract clauses; provides an agreement between two or more parties and asks to identify possible disputed points; describes performance deviations that have occurred and needs the legal consequences assessed; forward-looking risk identification in contract review, due diligence, or deal negotiation. The skill focuses on two core questions — \"will a legal dispute arise\" and \"is there breach risk\" — and outputs a structured risk list with response recommendations." }, "es": { "name": "dispute-and-performance-risk", "summary": "Se activa cuando el usuario envía textos contractuales, esquemas de transacción, acuerdos comerciales o descripciones de relaciones jurídicas y necesita evaluar posibles disputas legales o riesgos de incumplimiento dentro de ellos. Escenarios típicos de activación incluyen, entre otros: el usuario pide revisar un contrato y señalar riesgos potenciales; pregunta si un esquema de transacción enfrenta obstáculos de cumplimiento; describe dificultades en la ejecución del contrato y necesita determinar si constituyen incumplimiento; pide una calificación de riesgo de las cláusulas contractuales; aporta un acuerdo entre dos o más partes y pide identificar posibles puntos de disputa; describe desviaciones de cumplimiento ya ocurridas y necesita evaluar las consecuencias jurídicas; identificación prospectiva de riesgos en revisión de contratos, due diligence o negociación de transacciones. La habilidad se centra en dos preguntas nucleares —\"¿surgirá una disputa legal?\" y \"¿existe riesgo de incumplimiento?\"— y produce una lista estructurada de riesgos con recomendaciones de respuesta." }, "fr": { "name": "dispute-and-performance-risk", "summary": "Se déclenche lorsque l'utilisateur soumet des textes contractuels, des montages transactionnels, des accords commerciaux ou des descriptions de relations juridiques et doit évaluer les litiges juridiques ou risques d'inexécution potentiels qu'ils comportent. Scénarios types de déclenchement, sans s'y limiter : l'utilisateur demande de réviser un contrat et de signaler les risques potentiels ; demande si un montage transactionnel comporte des obstacles à l'exécution ; décrit des difficultés d'exécution du contrat et doit déterminer s'il y a inexécution ; demande une notation des risques des clauses contractuelles ; fournit un accord entre deux parties ou plus et demande d'identifier les points de litige possibles ; décrit des écarts d'exécution survenus et fait évaluer leurs conséquences juridiques ; identification prospective des risques lors de la revue de contrats, de la due diligence ou de la négociation. La compétence se concentre sur deux questions centrales — « un litige surviendra-t-il ? » et « y a-t-il un risque d'inexécution ? » — et produit une liste structurée de risques avec des recommandations de réponse." }, "ja": { "name": "dispute-and-performance-risk", "summary": "ユーザーが契約書テキスト、取引スキーム、商業契約、法律関係の記述を提出し、そこから生じ得る法的紛争や債務不履行リスクを評価する必要があるときに起動します。典型的な起動場面（これらに限りません）：契約を審査して潜在的リスクを指摘してほしい；ある取引スキームに履行障害があるか知りたい；契約履行中の困難を説明し、債務不履行に当たるか判断してほしい；契約条項のリスク評価を求める；二者以上の合意書から争点を特定してほしい；すでに発生した履行の逸脱について法的結果を評価してほしい；契約審査、デューデリジェンス、取引交渉などの場面での前向きなリスク特定。本スキルは「法的紛争が生じるか」「債務不履行リスクがあるか」の2つの核心的問題に焦点を当て、構造化されたリスク一覧と対応提案を出力します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:dispute-issue-identification", "slug": "skill-dispute-issue-identification", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Disputed-issue identification: after legal element extraction, it extracts the disputed issues from the legal facts and further converts case relationships into legal-relation questions. It can exclude uncontested matters and effectively focus subsequent analysis on the parties' contested facts, evi" }, "human": { "zh-CN": { "name": "dispute-issue-identification", "summary": "争议焦点识别是指在法律要素提取之后，从法律事实中提取争议焦点，进一步将案件关系转化为法律关系问题。 能够排除无争议事项，能够有效围绕当事人争议的事实、证据和法律适用等焦点问题进行后续分析。 通过事实调查、证据审查等梳理再次确定争议焦点，可更加深入和靠近争议焦点的本质，将案件剖析的更加充分和透彻。" }, "en": { "name": "dispute-issue-identification", "summary": "Disputed-issue identification: after legal element extraction, it extracts the disputed issues from the legal facts and further converts case relationships into legal-relation questions. It can exclude uncontested matters and effectively focus subsequent analysis on the parties' contested facts, evidence, and application of law. By re-determining the disputed issues through fact investigation and evidence review, it gets closer to the essence of each issue and dissects the case more fully and thoroughly." }, "es": { "name": "dispute-issue-identification", "summary": "Identificación de puntos de disputa: tras la extracción de elementos jurídicos, extrae los puntos controvertidos de los hechos jurídicos y convierte aún más las relaciones del caso en cuestiones de relaciones jurídicas. Puede excluir los asuntos no controvertidos y enfocar eficazmente el análisis posterior en los hechos disputados por las partes, la prueba y la aplicación del derecho. Al redeterminar los puntos de disputa mediante investigación de hechos y examen de pruebas, se acerca más a su esencia y diseca el caso de manera más completa y exhaustiva." }, "fr": { "name": "dispute-issue-identification", "summary": "Identification des points litigieux : après l'extraction des éléments juridiques, elle extrait les points litigieux des faits juridiques et convertit ensuite les relations du dossier en questions de relations juridiques. Elle peut écarter les éléments non contestés et concentrer efficacement l'analyse ultérieure sur les faits contestés par les parties, les preuves et l'application du droit. En redéterminant les points litigieux par l'enquête factuelle et l'examen des preuves, elle s'approche davantage de leur essence et dissèque le dossier de façon plus complète et plus approfondie." }, "ja": { "name": "dispute-issue-identification", "summary": "争点識別とは、法律要素の抽出後に法律事実から争点を抽出し、さらに事件関係を法律関係の問題へと転換することです。争いのない事項を除外でき、当事者が争う事実、証拠、法律適用などの焦点的事項に効果的に後続分析を絞り込めます。事実調査や証拠審査を通じて争点を改めて確定することで、争点の本質により深く迫り、事件をより十分かつ徹底的に剖析できます。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:evidence-argument-chain", "slug": "skill-evidence-argument-chain", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when multiple pieces of evidence in a case need to be systematically mapped to legal argument claims. Typical trigger scenarios include: building complete evidence-to-claim correspondences when preparing litigation documents (complaints, statements of defense, statements of representation);" }, "human": { "zh-CN": { "name": "evidence-argument-chain", "summary": "当需要将案件中的多项证据与法律论证主张进行系统性对应挂钩时触发此技能。 典型触发场景包括：准备诉讼文书（起诉状、答辩状、代理词）时需要构建完整的证据-主张对应关系； 庭审前整理证据目录并明确每项证据的证明目的；对方提出反驳后需要重新组织证据链以回应； 案件涉及多个争议焦点，需要将证据分组归类到不同论证主张下； 需要评估现有证据是否足以支撑全部主张，识别证据链中的薄弱环节和缺口。 本技能的核心是建立\"主张→要件→证据→证明力评估\"的完整映射关系，确保每个法律主张都有充分的证据支撑， 每项证据都有明确的证明目的，整体论证链条逻辑严密、层次清晰。" }, "en": { "name": "evidence-argument-chain", "summary": "Triggers when multiple pieces of evidence in a case need to be systematically mapped to legal argument claims. Typical trigger scenarios include: building complete evidence-to-claim correspondences when preparing litigation documents (complaints, statements of defense, statements of representation); organizing the evidence list before trial and clarifying each item's probative purpose; reorganizing the evidence chain to respond after the opponent rebuts; grouping evidence under different argument claims when a case involves multiple disputed issues; assessing whether the existing evidence suffices to support all claims, and identifying weak links and gaps in the evidence chain. The core is establishing the complete mapping of \"claim → element → evidence → probative-force assessment\", ensuring every legal claim has sufficient evidentiary support and every item of evidence a clear probative purpose, with the overall argument chain logically tight and clearly layered." }, "es": { "name": "evidence-argument-chain", "summary": "Se activa cuando múltiples pruebas del caso deben vincularse sistemáticamente con las pretensiones argumentativas jurídicas. Escenarios típicos de activación: construir correspondencias completas prueba-pretensión al preparar escritos litigiosos (demanda, contestación, alegatos de representación); organizar el listado de pruebas antes del juicio y precisar el fin probatorio de cada elemento; reorganizar la cadena probatoria para responder tras la réplica de la contraparte; agrupar pruebas bajo distintas pretensiones cuando el caso involucra varios puntos de disputa; evaluar si las pruebas existentes bastan para sostener todas las pretensiones e identificar eslabones débiles y vacíos en la cadena probatoria. El núcleo es establecer la correspondencia completa \"pretensión → elemento → prueba → valoración de la fuerza probatoria\", garantizando que cada pretensión jurídica tenga respaldo probatorio suficiente y cada prueba un fin probatorio claro, con una cadena argumentativa lógicamente sólida y claramente estratificada." }, "fr": { "name": "evidence-argument-chain", "summary": "Se déclenche lorsque plusieurs preuves du dossier doivent être systématiquement reliées aux thèses argumentatives juridiques. Scénarios types de déclenchement : construire des correspondances complètes preuve-thèse lors de la préparation des écritures (conclusions de poursuite, conclusions en défense, conclusions de représentation) ; organiser la liste des preuves avant l'audience et préciser le but probatoire de chaque pièce ; réorganiser la chaîne de preuves pour répondre après la réfutation adverse ; regrouper les preuves sous différentes thèses quand le dossier comporte plusieurs points litigieux ; évaluer si les preuves existantes suffisent à étayer toutes les thèses, et repérer les maillons faibles et les lacunes de la chaîne probatoire. Le cœur du dispositif : établir la cartographie complète « thèse → élément → preuve → évaluation de la force probatoire », garantissant que chaque thèse juridique bénéficie d'un soutien probatoire suffisant et chaque preuve d'un but probatoire clair, pour une chaîne argumentative logiquement serrée et clairement hiérarchisée." }, "ja": { "name": "evidence-argument-chain", "summary": "事件内の複数の証拠と法的論証の主張を体系的に対応付ける必要があるときに起動します。典型的な起動場面：訴訟文書（訴状、答弁書、代理人意見書）の準備時に証拠-主張の対応関係を構築する；公判前に証拠リストを整理し、各証拠の証明目的を明確にする；相手方の反論を受け、応答するために証拠連鎖を再編成する；事件に複数の争点があり、証拠を異なる論証主張の下にグループ化する；既存の証拠がすべての主張を支えるのに十分か評価し、証拠連鎖の弱点と欠落を特定する。本スキルの核心は「主張→要件→証拠→証明力評価」の完全なマッピングを確立し、すべての法的主張に十分な証拠支持が、すべての証拠に明確な証明目的があることを保証し、論証連鎖全体を論理的に緻密で階層の明確なものにすることです。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:evidence-evaluation", "slug": "skill-evidence-evaluation", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when the case's evidence materials need assessment of authenticity, legality, and relevance (the \"three properties\" of evidence) and of probative force. Typical trigger scenarios include but are not limited to: after a party or agent submits evidence, judging whether the court can admit it;" }, "human": { "zh-CN": { "name": "evidence-evaluation", "summary": "当需要对案件中的证据材料进行真实性、合法性、关联性（证据\"三性\"）及证明力评估时触发此技能。 典型触发场景包括但不限于： - 当事人或代理人提交证据后，需要判断该证据能否被法庭采纳 - 对方当事人对己方证据提出质证意见，需要进行回应和分析 - 需要对全案证据进行综合评判，确定待证事实是否成立 - 需要判断证据是否达到相应的证明标准（如高度盖然性、排除合理怀疑等） - 需要识别证据链中的薄弱环节并提出补强建议 - 电子数据、鉴定意见、证人证言等特殊证据形式的效力判断 - 非法证据排除的识别与论证" }, "en": { "name": "evidence-evaluation", "summary": "Triggers when the case's evidence materials need assessment of authenticity, legality, and relevance (the \"three properties\" of evidence) and of probative force. Typical trigger scenarios include but are not limited to: after a party or agent submits evidence, judging whether the court can admit it; responding to and analyzing the opposing party's cross-examination objections to one's own evidence; comprehensively evaluating all the case's evidence to determine whether the facts to be proven stand; judging whether the evidence meets the applicable proof standard (e.g., high preponderance, exclusion of reasonable doubt); identifying weak links in the evidence chain and proposing corroboration suggestions; determining the validity of special evidence forms such as electronic data, expert opinions, and witness testimony; recognizing and arguing the exclusion of illegally obtained evidence." }, "es": { "name": "evidence-evaluation", "summary": "Se activa cuando hay que valorar la autenticidad, legalidad y pertinencia (las \"tres propiedades\" de la prueba) y la fuerza probatoria del material probatorio del caso. Escenarios típicos de activación incluyen, entre otros: tras la presentación de pruebas por una parte o su representante, determinar si el tribunal puede admitirlas; responder y analizar las objeciones de la contraparte sobre las pruebas propias; valorar de forma integral todas las pruebas del caso para determinar si los hechos a probar quedan acreditados; juzgar si la prueba alcanza el estándar probatorio aplicable (p. ej., alta probabilidad, exclusión de la duda razonable); identificar eslabones débiles de la cadena probatoria y proponer refuerzos; determinar la validez de formas probatorias especiales como datos electrónicos, dictámenes periciales y testimonios; identificar y argumentar la exclusión de pruebas ilícitas." }, "fr": { "name": "evidence-evaluation", "summary": "Se déclenche lorsqu'il faut évaluer l'authenticité, la licéité et la pertinence (les « trois propriétés » de la preuve) ainsi que la force probatoire des pièces du dossier. Scénarios types de déclenchement, sans s'y limiter : après la production de preuves par une partie ou son mandataire, déterminer si le tribunal peut les admettre ; répondre aux observations de la partie adverse sur ses propres preuves ; évaluer globalement l'ensemble des preuves du dossier pour déterminer si les faits à prouver sont établis ; juger si la preuve atteint le standard probatoire applicable (ex. : haute probabilité, exclusion du doute raisonnable) ; repérer les maillons faibles de la chaîne probatoire et proposer des renforts ; apprécier la validité des formes de preuve particulières (données électroniques, expertise, témoignages) ; identifier et argumenter l'exclusion des preuves illicites." }, "ja": { "name": "evidence-evaluation", "summary": "事件の証拠資料について、真実性・合法性・関連性（証拠の「三性」）および証明力を評価する必要があるときに起動します。典型的な起動場面（これらに限りません）：当事者や代理人が証拠を提出した後、法廷が採用できるか判断する；相手方から己方の証拠への質証意見が出され、応答・分析する必要がある；全事件の証拠を総合評価し、立証すべき事実が成立するか確定する；証拠が対応する証明基準（高度な蓋然性、合理的疑いの排除など）に達するか判断する；証拠連鎖の弱点を特定し補強案を提案する；電子データ、鑑定意見、証人証言など特殊な証拠形式の効力判断；違法収集証拠排除の特定と論証。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:formal-legal-consequence", "slug": "skill-formal-legal-consequence", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when specific legal consequences must be derived from confirmed legal facts, identified legal relations, and matched legal norms — including but not limited to the type and scope of civil liability, damages calculation, criminal sentence determination, administrative penalty types and range" }, "human": { "zh-CN": { "name": "formal-legal-consequence", "summary": "当需要根据已确认的法律事实、已识别的法律关系和已匹配的法律规范，推导出具体的法律后果（包括但不限于民事责任类型与范围、损害赔偿金额计算、刑事刑期确定、行政处罚种类与幅度、合同效力状态等）时，触发本技能。 典型触发场景包括： - 用户询问\"我能获得多少赔偿\"\"对方要承担什么责任\" - 用户询问\"这个行为会判几年\"\"罚款多少\" - 需要从法律构成要件满足情况推导出具体法律效果 - 需要计算具体赔偿数额、刑期范围、违约金数额等量化结果 - 需要综合考虑加重、减轻、免除等调整因素后得出最终法律后果 本技能是法律推理链条的终端环节，将抽象的法律规范转化为具体的、可操作的法律结论。" }, "en": { "name": "formal-legal-consequence", "summary": "Triggers when specific legal consequences must be derived from confirmed legal facts, identified legal relations, and matched legal norms — including but not limited to the type and scope of civil liability, damages calculation, criminal sentence determination, administrative penalty types and ranges, and contract validity status. Typical trigger scenarios include: the user asks \"how much compensation can I get\" or \"what liability will the other party bear\"; asks \"how many years will this act draw\" or \"how big is the fine\"; deriving concrete legal effects from the satisfaction of legal constitutive elements; computing quantified results such as compensation amounts, sentence ranges, and liquidated damages amounts; weighing aggravating, mitigating, and exempting factors to reach the final legal consequence. This skill is the terminal link of the legal reasoning chain, converting abstract legal norms into concrete, actionable legal conclusions." }, "es": { "name": "formal-legal-consequence", "summary": "Se activa cuando deben derivarse consecuencias jurídicas concretas a partir de los hechos jurídicos confirmados, las relaciones jurídicas identificadas y las normas jurídicas emparejadas —incluyendo, entre otras, el tipo y alcance de la responsabilidad civil, el cálculo de la indemnización, la determinación de la pena penal, los tipos y rangos de sanciones administrativas y el estado de validez del contrato. Escenarios típicos de activación: el usuario pregunta \"¿cuánta indemnización puedo obtener?\" o \"¿qué responsabilidad asumirá la otra parte?\"; pregunta \"¿cuántos años acarrea esta conducta?\" o \"¿de cuánto es la multa?\"; derivar efectos jurídicos concretos de la satisfacción de los elementos constitutivos legales; calcular resultados cuantificados como importes de indemnización, rangos de pena e importes de cláusula penal; ponderar factores agravantes, atenuantes y eximentes hasta llegar a la consecuencia jurídica final. Esta habilidad es el eslabón terminal de la cadena de razonamiento jurídico: convierte normas jurídicas abstractas en conclusiones jurídicas concretas y accionables." }, "fr": { "name": "formal-legal-consequence", "summary": "Se déclenche lorsque des conséquences juridiques concrètes doivent être déduites des faits juridiques confirmés, des relations juridiques identifiées et des normes juridiques appariées — y compris, sans s'y limiter, le type et l'étendue de la responsabilité civile, le calcul de l'indemnisation, la détermination de la peine pénale, les types et plages de sanctions administratives, et l'état de validité du contrat. Scénarios types de déclenchement : l'utilisateur demande « quelle indemnisation puis-je obtenir ? » ou « quelle responsabilité l'autre partie encourt-elle ? » ; demande « combien d'années cet acte vaut-il ? » ou « quelle est l'amende ? » ; déduire des effets juridiques concrets de la satisfaction des éléments constitutifs légaux ; calculer des résultats quantifiés comme les montants d'indemnisation, les plages de peine et les montants de clause pénale ; pondérer les facteurs aggravants, atténuants et exonératoires jusqu'à la conséquence juridique finale. Cette compétence est le maillon terminal de la chaîne du raisonnement juridique : elle transforme des normes juridiques abstraites en conclusions juridiques concrètes et opérationnelles." }, "ja": { "name": "formal-legal-consequence", "summary": "確定した法律事実、識別済みの法律関係、マッチ済みの法律規範から、具体的な法的帰結（民事責任の類型と範囲、損害賠償額の計算、刑事刑期の確定、行政処分の種類と幅度、契約の効力状態など。これらに限りません）を導く必要があるときに起動します。典型的な起動場面：ユーザーが「いくら賠償を受けられるか」「相手はどんな責任を負うか」と尋ねる；「この行為は何年の刑になるか」「罰金はいくらか」と尋ねる；法律構成要件の充足状況から具体的な法的効果を導く；賠償額、刑期範囲、違約金額などの定量的結果を計算する；加重・減軽・免除などの調整要素を総合して最終的な法的帰結を導く。本スキルは法律推論連鎖の末端リンクであり、抽象的な法律規範を具体的で実行可能な法的結論へ変換します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:inductive-reasoning", "slug": "skill-inductive-reasoning", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when general legal rules, adjudication rules, or legal principles need to be distilled and summarized from one or more specific cases, judgments, or fact patterns. Typical trigger scenarios include: the user asks to induce the gist of rulings from a series of cases, extract generalizable le" }, "human": { "zh-CN": { "name": "inductive-reasoning", "summary": "当需要从一个或多个具体案例、判决、事实模式中提炼、总结出一般性法律规则、裁判规则或法律原则时，触发此技能。典型触发场景包括：用户要求从系列案例中归纳裁判要旨、从个案中提炼可推广的法律命题、从多个事实模式中识别共同法律规则、构建类案检索的裁判规则摘要、以及从实务判例中总结司法适用标准。" }, "en": { "name": "inductive-reasoning", "summary": "Triggers when general legal rules, adjudication rules, or legal principles need to be distilled and summarized from one or more specific cases, judgments, or fact patterns. Typical trigger scenarios include: the user asks to induce the gist of rulings from a series of cases, extract generalizable legal propositions from individual cases, identify common legal rules across multiple fact patterns, build adjudication-rule summaries for similar-case retrieval, and summarize judicial application standards from practice precedents." }, "es": { "name": "inductive-reasoning", "summary": "Se activa cuando hay que destilar y resumir reglas jurídicas generales, criterios de decisión o principios jurídicos a partir de uno o varios casos, sentencias o patrones fácticos concretos. Escenarios típicos de activación: el usuario pide inducir la doctrina esencial de una serie de casos, extraer de casos individuales proposiciones jurídicas generalizables, identificar reglas jurídicas comunes en varios patrones fácticos, construir resúmenes de criterios de decisión para la búsqueda de casos análogos, y resumir estándares de aplicación judicial a partir de precedentes de la práctica." }, "fr": { "name": "inductive-reasoning", "summary": "Se déclenche lorsqu'il faut dégager et résumer des règles juridiques générales, des règles de jugement ou des principes juridiques à partir d'un ou plusieurs cas, jugements ou schémas factuels concrets. Scénarios types de déclenchement : l'utilisateur demande d'induire les motifs essentiels d'une série d'affaires, d'extraire de cas individuels des propositions juridiques généralisables, d'identifier des règles juridiques communes à plusieurs schémas factuels, de construire des résumés de règles de jugement pour la recherche d'affaires analogues, et de synthétiser les standards d'application judiciaire à partir de précédents de la pratique." }, "ja": { "name": "inductive-reasoning", "summary": "一つ以上の具体的な事例、判決、事実パターンから一般法的ルール、裁判規則、法原則を抽出・要約する必要があるときに起動します。典型的な起動場面：一連の事例から裁判要旨を帰納する、個別事件から一般化可能な法的命題を抽出する、複数の事実パターンから共通の法的ルールを識別する、類似事件検索用の裁判規則サマリーを構築する、実務判例から司法適用基準をまとめる、など。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:internal-compliance-risk-identification", "slug": "skill-internal-compliance-risk-identification", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when the enterprise's internal compliance management system needs a systematic review to identify policy gaps, process defects, and data-privacy compliance risks. Typical trigger scenarios include but are not limited to: the enterprise commissions lawyers or compliance consultants for an in" }, "human": { "zh-CN": { "name": "internal-compliance-risk-identification", "summary": "当需要对企业内部的合规管理体系进行系统性审查，识别制度缺漏、流程缺陷和数据隐私合规风险时触发此技能。 典型触发场景包括但不限于： - 企业委托律师或合规顾问进行内部合规风险评估 - 企业拟上市、并购或接受监管检查前进行合规自查 - 发生合规事件后，需要排查管理体系中的系统性缺陷 - 企业建立或更新合规管理体系时，需要评估现有制度与法律法规的匹配度 - 用户要求审查企业内部规章制度、核心业务流程或隐私政策协议的合规性 本技能覆盖三大审查维度：制度体系完整性审查、业务流程控制有效性审查、个人信息保护合规性审查。" }, "en": { "name": "internal-compliance-risk-identification", "summary": "Triggers when the enterprise's internal compliance management system needs a systematic review to identify policy gaps, process defects, and data-privacy compliance risks. Typical trigger scenarios include but are not limited to: the enterprise commissions lawyers or compliance consultants for an internal compliance risk assessment; compliance self-checks before a planned IPO, M&A deal, or regulatory inspection; investigating systemic defects in the management system after a compliance incident; assessing the fit between existing policies and laws and regulations when building or updating the compliance management system; the user asks to review the compliance of internal rules, core business processes, or privacy policy agreements. The skill covers three review dimensions: completeness review of the policy system, effectiveness review of business process controls, and compliance review of personal-information protection." }, "es": { "name": "internal-compliance-risk-identification", "summary": "Se activa cuando el sistema de gestión de cumplimiento interno de la empresa requiere una revisión sistemática para identificar lagunas normativas, defectos de proceso y riesgos de cumplimiento de la privacidad de datos. Escenarios típicos de activación incluyen, entre otros: la empresa encarga a abogados o consultores de cumplimiento una evaluación interna de riesgos de cumplimiento; autoevaluaciones antes de una salida a bolsa, una fusión/adquisición o una inspección regulatoria; investigación de defectos sistémicos del sistema de gestión tras un incidente de cumplimiento; evaluar el ajuste entre las políticas vigentes y las leyes y reglamentos al construir o actualizar el sistema de gestión de cumplimiento; el usuario pide revisar la conformidad de las normas internas, los procesos de negocio nucleares o los acuerdos de política de privacidad. La habilidad cubre tres dimensiones de revisión: integridad del sistema normativo, efectividad de los controles de procesos de negocio y conformidad de la protección de datos personales." }, "fr": { "name": "internal-compliance-risk-identification", "summary": "Se déclenche lorsque le système de gestion de la conformité interne de l'entreprise doit faire l'objet d'une revue systématique afin d'identifier les lacunes des règles, les défauts de processus et les risques de conformité de la confidentialité des données. Scénarios types de déclenchement, sans s'y limiter : l'entreprise mandate des avocats ou des consultants conformité pour une évaluation interne des risques de conformité ; autocontrôles avant une introduction en bourse, une fusion-acquisition ou une inspection réglementaire ; investigation des défauts systémiques du système de gestion après un incident de conformité ; évaluation de l'adéquation entre les règles existantes et les lois et règlements lors de la construction ou de la mise à jour du système de gestion de la conformité ; l'utilisateur demande de vérifier la conformité des règles internes, des processus métier essentiels ou des accords de politique de confidentialité. La compétence couvre trois dimensions de revue : complétude du système de règles, efficacité des contrôles des processus métier et conformité de la protection des données personnelles." }, "ja": { "name": "internal-compliance-risk-identification", "summary": "企業内部のコンプライアンス管理体系を体系的に審査し、制度の欠落、プロセスの欠陥、データプライバシー・コンプライアンスリスクを特定する必要があるときに起動します。典型的な起動場面（これらに限りません）：企業が弁護士やコンプライアンス顧問に内部コンプライアンスリスク評価を委託する；上場、M&A、規制検査の前にコンプライアンス自己点検を行う；コンプライアンス事案発生後に管理体系のシステミックな欠陥を調査する；コンプライアンス管理体系の構築・更新時に、既存制度と法令との整合性を評価する；ユーザーが社内規程、基幹業務プロセス、プライバシーポリシー契約の適合性審査を求める。本スキルは3つの審査次元をカバーします：制度体系の完全性審査、業務プロセス管理の有効性審査、個人情報保護の適合性審査。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:judgment-document-generation", "slug": "skill-judgment-document-generation", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Based on a description of case facts, coordinates and invokes 8 atomic capabilities to generate a complete criminal judgment with standard formatting and rigorous argumentation. Covers the full workflow of fact-finding, application of law, sentencing reasoning, and document output." }, "human": { "zh-CN": { "name": "judgment-document-generation", "summary": "根据案件事实描述，协调调用8个原子能力，生成格式规范、论证严密的完整刑事判决书。涵盖事实认定、法律适用、量刑推理及文书输出全流程。" }, "en": { "name": "judgment-document-generation", "summary": "Based on a description of case facts, coordinates and invokes 8 atomic capabilities to generate a complete criminal judgment with standard formatting and rigorous argumentation. Covers the full workflow of fact-finding, application of law, sentencing reasoning, and document output." }, "es": { "name": "judgment-document-generation", "summary": "A partir de la descripción de los hechos del caso, coordina e invoca 8 capacidades atómicas para generar una sentencia penal completa, con formato normalizado y argumentación rigurosa. Cubre todo el flujo: determinación de hechos, aplicación del derecho, razonamiento de la pena y salida del documento." }, "fr": { "name": "judgment-document-generation", "summary": "À partir d'une description des faits du dossier, coordonne et invoque 8 capacités atomiques pour générer un jugement pénal complet, au format normalisé et à l'argumentation rigoureuse. Couvre tout le flux : constatation des faits, application du droit, raisonnement sur la peine et production du document." }, "ja": { "name": "judgment-document-generation", "summary": "事件事実の記述に基づき、8つのアトミック能力を調整・呼び出して、書式が整い論証の厳密な完全な刑事判決書を生成します。事実認定、法律適用、量刑推論、文書出力の全プロセスをカバーします。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:judicial-value-judgment", "slug": "skill-judicial-value-judgment", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Assists judges in making structured, reviewable, and arguable judicial value judgments in scenarios of rights conflicts, legal uncertainty, proportionality review, and discretionary reasoning. This Skill assists judges in difficult cases to identify legal value conflicts, determine the boundaries of" }, "human": { "zh-CN": { "name": "judicial-value-judgment", "summary": "辅助法官在权利冲突、法律不确定、比例原则审查和裁量说理场景中，进行结构化、可审查、可论证的司法价值判断。 本 Skill 用于辅助法官在疑难案件中识别法律价值冲突、确定裁量边界、展开价值衡量，并形成可用于裁判说理的分析框架，不替代法官作出最终裁判，而是帮助法官将价值判断转化为有事实基础、规范基础和论证结构的司法推理过程。" }, "en": { "name": "judicial-value-judgment", "summary": "Assists judges in making structured, reviewable, and arguable judicial value judgments in scenarios of rights conflicts, legal uncertainty, proportionality review, and discretionary reasoning. This Skill assists judges in difficult cases to identify legal value conflicts, determine the boundaries of discretion, conduct value weighing, and form an analytical framework usable for decision reasoning; it does not replace the judge in making the final ruling, but helps the judge convert value judgments into a judicial reasoning process with factual grounding, normative grounding, and argumentative structure." }, "es": { "name": "judicial-value-judgment", "summary": "Asiste a los jueces en la formación de juicios de valor judiciales estructurados, revisables y argumentables en escenarios de conflicto de derechos, incertidumbre jurídica, examen de proporcionalidad y motivación de la discrecionalidad. Este Skill ayuda al juez, en los casos difíciles, a identificar conflictos de valores jurídicos, determinar los límites de la discrecionalidad, realizar la ponderación de valores y construir un marco analítico utilizable para la motivación de la decisión; no sustituye al juez en la decisión final, sino que ayuda a convertir el juicio de valor en un proceso de razonamiento judicial con base fáctica, base normativa y estructura argumentativa." }, "fr": { "name": "judicial-value-judgment", "summary": "Aide les juges à porter des jugements de valeur judiciaires structurés, contrôlables et argumentables dans les situations de conflit de droits, d'incertitude juridique, de contrôle de proportionnalité et de motivation de l'appréciation. Cette Skill aide le juge, dans les dossiers difficiles, à identifier les conflits de valeurs juridiques, à délimiter les bornes du pouvoir d'appréciation, à mener la pondération des valeurs et à construire un cadre d'analyse exploitable pour la motivation de la décision ; elle ne remplace pas le juge dans la décision finale, mais l'aide à convertir le jugement de valeur en un processus de raisonnement judiciaire doté d'une base factuelle, d'une base normative et d'une structure argumentative." }, "ja": { "name": "judicial-value-judgment", "summary": "権利衝突、法の不確実性、比例原則の審査、裁量の理由付けといった場面で、裁判官が構造化され、検証可能で論証可能な司法的価値判断を行えるよう支援します。本Skillは、難しい事件において裁判官が法的価値の衝突を識別し、裁量の限界を確定し、価値衡量を展開し、裁判理由付けに使える分析フレームワークを形成するのを支援します。裁判官の最終判断を代替するものではなく、価値判断を、事実的基盤・規範的基盤・論証構造を備えた司法推論プロセスへと転換する助けとなるものです。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:legal-article-retrieval", "slug": "skill-legal-article-retrieval", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Generates standardized legal research reports, suitable for case retrieval, statute retrieval, similar-case analysis, and other legal research scenarios. Use when the user needs to perform industry or professional analysis, predict adjudication outcomes, re-verify the validity of legal bases, confir" }, "human": { "zh-CN": { "name": "legal-article-retrieval", "summary": "生成标准化法律检索报告，适用于案例检索、法条检索、类案分析等法律研究场景。当用户需要进行行业或专业分析、预测裁判结果、复核法律依据的有效性、确认请求、抗辩主张的法律依据、分析司法实践倾向等任务时使用。支持诉讼方案评估、庭审准备、合规审查、法律培训等各类法律实务场景。" }, "en": { "name": "legal-article-retrieval", "summary": "Generates standardized legal research reports, suitable for case retrieval, statute retrieval, similar-case analysis, and other legal research scenarios. Use when the user needs to perform industry or professional analysis, predict adjudication outcomes, re-verify the validity of legal bases, confirm the legal grounds of claims or defenses, or analyze tendencies in judicial practice. Supports litigation plan evaluation, trial preparation, compliance review, legal training, and other legal practice scenarios." }, "es": { "name": "legal-article-retrieval", "summary": "Genera informes normalizados de investigación jurídica, aptos para la búsqueda de casos, la búsqueda de preceptos legales, el análisis de casos análogos y otros escenarios de investigación jurídica. Use cuando el usuario necesite realizar análisis sectoriales o profesionales, prever los resultados de la decisión judicial, reverificar la vigencia de los fundamentos legales, confirmar las bases jurídicas de pretensiones o defensas, o analizar las tendencias de la práctica judicial. Apoya la evaluación de planes litigiosos, la preparación del juicio, la revisión de cumplimiento, la formación jurídica y otros escenarios de práctica jurídica." }, "fr": { "name": "legal-article-retrieval", "summary": "Génère des rapports normalisés de recherche juridique, adaptés à la recherche d'affaires, à la recherche de dispositions légales, à l'analyse d'affaires analogues et à d'autres scénarios de recherche juridique. À utiliser lorsque l'utilisateur doit mener une analyse sectorielle ou professionnelle, prédire l'issue du jugement, revérifier la validité des fondements juridiques, confirmer les bases juridiques de prétentions ou de défenses, ou analyser les tendances de la pratique judiciaire. Prend en charge l'évaluation des plans contentieux, la préparation de l'audience, la revue de conformité, la formation juridique et d'autres scénarios de pratique juridique." }, "ja": { "name": "legal-article-retrieval", "summary": "標準化された法律リサーチレポートを生成します。事例検索、条文検索、類似事件分析などの法律リサーチ場面に適しています。ユーザーが業界・専門分析、裁判結果の予測、法的根拠の有効性の再確認、請求・抗弁の法的根拠の確認、司法実務の傾向分析などのタスクを行う必要があるときに使用します。訴訟方針の評価、公判準備、コンプライアンス審査、法律研修など、各種の法律実務場面をサポートします。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:legal-concept-comprehension", "slug": "skill-legal-concept-comprehension", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when the user asks to explain, distinguish, break down, or understand a legal concept. Typical trigger scenarios include but are not limited to: the user directly asks \"what is XX\" or \"how to understand XX\"; asks to compare the similarities and differences of two legal concepts (e.g., \"the " }, "human": { "zh-CN": { "name": "legal-concept-comprehension", "summary": "当用户要求解释、辨析、拆解或理解某个法律概念时触发此技能。 典型触发场景包括但不限于： - 用户直接询问\"什么是XX\"\"如何理解XX\" - 用户要求比较两个法律概念的异同（如\"不当得利与无因管理的区别\"） - 用户给出具体事实，询问是否构成某法律概念（如\"甲误将钱转给乙，是否构成不当得利？\"） - 用户在法律分析、文书写作、合同审查中遇到不熟悉的术语 - 用户要求拆解某概念的构成要件或分析其法律效果 本技能是法律分析的基础单元，为法律规则适用、案件分析、法律文书写作等提供概念基础。" }, "en": { "name": "legal-concept-comprehension", "summary": "Triggers when the user asks to explain, distinguish, break down, or understand a legal concept. Typical trigger scenarios include but are not limited to: the user directly asks \"what is XX\" or \"how to understand XX\"; asks to compare the similarities and differences of two legal concepts (e.g., \"the difference between unjust enrichment and negotiorum gestio\"); provides specific facts and asks whether they constitute a legal concept (e.g., \"A mistakenly transfers money to B — does this constitute unjust enrichment?\"); encounters unfamiliar terms during legal analysis, document drafting, or contract review; asks to break down a concept's constitutive elements or analyze its legal effects. This skill is the basic unit of legal analysis, providing the conceptual foundation for applying legal rules, case analysis, and legal document drafting." }, "es": { "name": "legal-concept-comprehension", "summary": "Se activa cuando el usuario pide explicar, distinguir, descomponer o comprender un concepto jurídico. Escenarios típicos de activación incluyen, entre otros: el usuario pregunta directamente \"¿qué es XX?\" o \"¿cómo entender XX?\"; pide comparar las semejanzas y diferencias de dos conceptos jurídicos (p. ej., \"la diferencia entre enriquecimiento injusto y gestión de negocios ajenos\"); aporta hechos concretos y pregunta si configuran un concepto jurídico (p. ej., \"A transfiere dinero por error a B, ¿constituye enriquecimiento injusto?\"); encuentra términos desconocidos durante el análisis jurídico, la redacción de documentos o la revisión de contratos; pide descomponer los elementos constitutivos de un concepto o analizar sus efectos jurídicos. Esta habilidad es la unidad básica del análisis jurídico y aporta la base conceptual para la aplicación de reglas jurídicas, el análisis de casos y la redacción de documentos jurídicos." }, "fr": { "name": "legal-concept-comprehension", "summary": "Se déclenche lorsque l'utilisateur demande d'expliquer, de distinguer, de décomposer ou de comprendre un concept juridique. Scénarios types de déclenchement, sans s'y limiter : l'utilisateur demande directement « qu'est-ce que XX ? » ou « comment comprendre XX ? » ; demande de comparer les similitudes et différences de deux concepts juridiques (ex. : « la différence entre l'enrichissement sans cause et la gestion d'affaires ») ; fournit des faits précis et demande s'ils constituent un concept juridique (ex. : « A transfère de l'argent par erreur à B, est-ce un enrichissement sans cause ? ») ; rencontre des termes inconnus lors de l'analyse juridique, de la rédaction ou de la revue de contrats ; demande de décomposer les éléments constitutifs d'un concept ou d'analyser ses effets juridiques. Cette compétence est l'unité de base de l'analyse juridique : elle fournit le socle conceptuel de l'application des règles juridiques, de l'analyse des dossiers et de la rédaction des documents juridiques." }, "ja": { "name": "legal-concept-comprehension", "summary": "ユーザーが法律概念の説明、弁別、分解、理解を求めるときに起動します。典型的な起動場面（これらに限りません）：「XXとは何か」「XXをどう理解すればいいか」と直接質問する；2つの法律概念の異同を比較してほしい（例：「不当利得と事務管理の違い」）；具体的な事実を提示して、ある法律概念を構成するか尋ねる（例：「Aが誤ってBに送金した場合、不当利得に当たるか？」）；法律分析、文書作成、契約審査の中で馴染みのない用語に出会う；ある概念の構成要件の分解や法律効果の分析を求める。本スキルは法律分析の基礎単元であり、法律規則の適用、事件分析、法律文書作成などに概念的基盤を提供します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:legal-document-formatting", "slug": "skill-legal-document-formatting", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Drafts complete civil or criminal judgments based on the adjudication-document standards of the Chinese People's Courts. Use when the user asks to \"draft a judgment\", \"generate an adjudication document\", \"write a civil judgment\", \"write a criminal judgment\", \"write the document from the case file\", " }, "human": { "zh-CN": { "name": "legal-document-formatting", "summary": "基于中国人民法院裁判文书制作规范，起草完整的民事判决书或刑事判决书。当用户要求\"起草判决书\"、\"生成裁判文书\"、\"写民事判决书\"、\"写刑事判决书\"、\"根据案卷写文书\"或提供案卷材料要求输出裁判文书时使用本技能，在起草各段落时自主调用原子技能（如争议焦点识别、法条检索、演绎推理、证据效力评估等）。" }, "en": { "name": "legal-document-formatting", "summary": "Drafts complete civil or criminal judgments based on the adjudication-document standards of the Chinese People's Courts. Use when the user asks to \"draft a judgment\", \"generate an adjudication document\", \"write a civil judgment\", \"write a criminal judgment\", \"write the document from the case file\", or provides case-file material requesting an adjudication document; while drafting each section, it autonomously invokes atomic skills (such as disputed-issue identification, statute retrieval, deductive reasoning, and evidence-validity assessment)." }, "es": { "name": "legal-document-formatting", "summary": "Redacta sentencias civiles o penales completas conforme a las normas de elaboración de documentos judiciales de los tribunales populares de China. Use cuando el usuario pida \"redactar una sentencia\", \"generar un documento judicial\", \"escribir una sentencia civil\", \"escribir una sentencia penal\", \"escribir el documento a partir del expediente\", o aporte material del expediente solicitando un documento judicial; al redactar cada sección invoca de forma autónoma habilidades atómicas (como la identificación de puntos de disputa, la búsqueda de preceptos, el razonamiento deductivo y la evaluación de la validez de la prueba)." }, "fr": { "name": "legal-document-formatting", "summary": "Rédige des jugements civils ou pénaux complets conformément aux normes de fabrication des documents judiciaires des tribunaux populaires de Chine. À utiliser lorsque l'utilisateur demande de « rédiger un jugement », de « générer un document judiciaire », d'« écrire un jugement civil », d'« écrire un jugement pénal », d'« écrire le document à partir du dossier », ou fournit le matériel du dossier en demandant un document judiciaire ; lors de la rédaction de chaque section, il invoque de manière autonome des compétences atomiques (telles que l'identification des points litigieux, la recherche de dispositions, le raisonnement déductif et l'évaluation de la validité des preuves)." }, "ja": { "name": "legal-document-formatting", "summary": "中国人民法院の裁判文書作成規範に基づき、完全な民事判決書または刑事判決書を起草します。ユーザーが「判決書を起草して」「裁判文書を生成して」「民事判決書を書いて」「刑事判決書を書いて」「事件記録から文書を書いて」と求めたときや、事件記録資料を提供して裁判文書の出力を求めたときに使用します。各段落の起草時に、争点識別、条文検索、演繹推論、証拠効力評価などのアトミックスキルを自律的に呼び出します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:legal-document-summarization", "slug": "skill-legal-document-summarization", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when structured summaries are needed for legal documents such as judgments, rulings, mediation agreements, arbitration awards, and administrative penalty decisions. Typical trigger scenarios include but are not limited to: the user asks to summarize a legal document, distill key points, or " }, "human": { "zh-CN": { "name": "legal-document-summarization", "summary": "当需要对判决书、裁定书、调解书、仲裁裁决书、行政处罚决定书等法律文书进行结构化摘要时触发此技能。 典型触发场景包括但不限于： - 用户要求对一份法律文书做摘要、提炼要点或生成裁判要旨 - 用户需要将多份法律文书转化为可对比、可检索的标准化摘要 - 用户要求提取案件信息、争议焦点、法律依据或说理过程 - 用户需要为案例研究、诉讼分析、教学备课准备摘要材料 - 用户要求区分不同文书类型（判决/裁定/调解/仲裁/行政处罚）的摘要重点 本技能的核心价值：在忠实于原文的基础上，将冗长法律文书转化为简明、客观、结构化的摘要，突出核心内容，避免复述全文和主观评价。" }, "en": { "name": "legal-document-summarization", "summary": "Triggers when structured summaries are needed for legal documents such as judgments, rulings, mediation agreements, arbitration awards, and administrative penalty decisions. Typical trigger scenarios include but are not limited to: the user asks to summarize a legal document, distill key points, or generate the gist of the ruling; needs to convert multiple legal documents into comparable, searchable standardized summaries; asks to extract case information, disputed issues, legal bases, or reasoning processes; needs summary material for case studies, litigation analysis, or teaching preparation; asks to differentiate summary emphases by document type (judgment/ruling/mediation/arbitration/administrative penalty). Core value of this skill: faithfully converting lengthy legal documents into concise, objective, structured summaries that highlight the core content, avoiding restating the full text and subjective commentary." }, "es": { "name": "legal-document-summarization", "summary": "Se activa cuando se necesitan resúmenes estructurados de documentos jurídicos como sentencias, providencias, actas de mediación, laudos arbitrales y resoluciones sancionadoras administrativas. Escenarios típicos de activación incluyen, entre otros: el usuario pide resumir un documento jurídico, destilar puntos clave o generar la doctrina esencial del fallo; necesita convertir varios documentos jurídicos en resúmenes normalizados comparables y consultables; pide extraer información del caso, puntos de disputa, fundamentos legales o procesos de razonamiento; necesita material de resumen para estudios de caso, análisis litigioso o preparación docente; pide diferenciar los énfasis del resumen según el tipo de documento (sentencia/providencia/mediación/arbitraje/sanción administrativa). Valor central de esta habilidad: convertir fielmente documentos jurídicos extensos en resúmenes concisos, objetivos y estructurados que resalten el contenido esencial, evitando repetir el texto íntegro y los juicios subjetivos." }, "fr": { "name": "legal-document-summarization", "summary": "Se déclenche lorsque des résumés structurés sont nécessaires pour des documents juridiques tels que jugements, ordonnances, accords de médiation, sentences arbitrales et décisions de sanction administrative. Scénarios types de déclenchement, sans s'y limiter : l'utilisateur demande de résumer un document juridique, d'en extraire les points clés ou de générer les motifs essentiels de la décision ; doit convertir plusieurs documents juridiques en résumés normalisés comparables et interrogeables ; demande d'extraire les informations du dossier, les points litigieux, les fondements juridiques ou les processus de raisonnement ; a besoin de matériel de synthèse pour des études de cas, une analyse contentieuse ou une préparation pédagogique ; demande de différencier les axes de synthèse selon le type de document (jugement/ordonnance/médiation/arbitrage/sanction administrative). Valeur centrale de cette compétence : convertir fidèlement de longs documents juridiques en résumés concis, objectifs et structurés qui mettent en avant le contenu essentiel, sans reprendre l'intégralité du texte ni d'appréciations subjectives." }, "ja": { "name": "legal-document-summarization", "summary": "判決書、裁定書、調書、仲裁判断書、行政処分決定書などの法律文書を構造化して要約する必要があるときに起動します。典型的な起動場面（これらに限りません）：法律文書の要約、要点の抽出、裁判要旨の生成を求める；複数の法律文書を比較可能で検索可能な標準化サマリーに変換する必要がある；事件情報、争点、法的根拠、説理過程の抽出を求める；事例研究、訴訟分析、授業準備のための要約資料が必要；文書タイプ別（判決/裁定/調停/仲裁/行政処分）に要約の重点を変えるよう求める。本スキルの核心的価値：原文に忠実な上で、長大な法律文書を簡潔・客観・構造化されたサマリーに変換し、核心内容を際立たせ、全文の繰り返しや主観的評価を避けることです。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:legal-element-extraction", "slug": "skill-legal-element-extraction", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when legally significant facts need to be extracted from unstructured text such as case descriptions, party statements, chat records, and media reports. Typical trigger scenarios include but are not limited to: the user provides a case description in everyday language and asks to extract th" }, "human": { "zh-CN": { "name": "legal-element-extraction", "summary": "当需要从案件描述、当事人陈述、聊天记录、媒体报道等非结构化文本中提取具有法律意义的事实时触发此技能。 典型触发场景包括但不限于： - 用户提供一段日常语言的案件描述，要求提取法律事实 - 用户要求将口语化、情绪化的叙事转化为法律分析可用的事实结构 - 用户要求区分案件叙述中的\"客观事实\"\"主观判断\"和\"法律评价\" - 用户要求将提取的事实对应到特定法律关系的构成要件 - 用户要求整理案件时间轴或制作事实梳理表 - 用户要求将生活语言转化为法律语言（如\"他一直拖着不还\"→\"履行期限届满后未履行\"） 本技能的核心价值：将混杂的生活叙事整理为可进入法律推理的事实结构，是法律分析的前置步骤。" }, "en": { "name": "legal-element-extraction", "summary": "Triggers when legally significant facts need to be extracted from unstructured text such as case descriptions, party statements, chat records, and media reports. Typical trigger scenarios include but are not limited to: the user provides a case description in everyday language and asks to extract the legal facts; asks to convert colloquial, emotional narratives into fact structures usable for legal analysis; asks to distinguish \"objective facts\", \"subjective judgments\", and \"legal evaluations\" in the case narrative; asks to map the extracted facts to the constitutive elements of a specific legal relation; asks to organize a case timeline or build a fact-sorting table; asks to translate everyday language into legal language (e.g., \"he keeps stalling on repayment\" → \"non-performance after the performance deadline has expired\"). Core value of this skill: organizing mixed everyday narratives into fact structures that can enter legal reasoning — a prerequisite step of legal analysis." }, "es": { "name": "legal-element-extraction", "summary": "Se activa cuando hay que extraer hechos jurídicamente relevantes de textos no estructurados como descripciones del caso, declaraciones de las partes, registros de chat y reportajes mediáticos. Escenarios típicos de activación incluyen, entre otros: el usuario aporta una descripción del caso en lenguaje cotidiano y pide extraer los hechos jurídicos; pide convertir narraciones coloquiales y emotivas en estructuras fácticas utilizables para el análisis jurídico; pide distinguir \"hechos objetivos\", \"juicios subjetivos\" y \"valoraciones jurídicas\" en la narración del caso; pide corresponder los hechos extraídos con los elementos constitutivos de una relación jurídica concreta; pide ordenar la cronología del caso o elaborar un cuadro de ordenación de hechos; pide convertir lenguaje cotidiano en lenguaje jurídico (p. ej., \"lleva tiempo dándole largas al pago\" → \"incumplimiento tras el vencimiento del plazo de cumplimiento\"). Valor central de esta habilidad: ordenar narraciones vitales heterogéneas en estructuras fácticas susceptibles de entrar en el razonamiento jurídico; es un paso previo del análisis jurídico." }, "fr": { "name": "legal-element-extraction", "summary": "Se déclenche lorsqu'il faut extraire des faits juridiquement significatifs de textes non structurés comme des descriptions de dossiers, des déclarations des parties, des échanges de messagerie ou des reportages. Scénarios types de déclenchement, sans s'y limiter : l'utilisateur fournit une description du dossier en langage courant et demande d'en extraire les faits juridiques ; demande de convertir des récits familiers et émotionnels en structures factuelles exploitables pour l'analyse juridique ; demande de distinguer « faits objectifs », « jugements subjectifs » et « évaluations juridiques » dans le récit du dossier ; demande de faire correspondre les faits extraits aux éléments constitutifs d'une relation juridique donnée ; demande d'établir une chronologie du dossier ou un tableau de tri des faits ; demande de traduire le langage de tous les jours en langage juridique (ex. : « il traîne pour rembourser » → « inexécution après l'expiration du délai d'exécution »). Valeur centrale de cette compétence : organiser des récits hétéroclites de la vie quotidienne en structures factuelles pouvant entrer dans le raisonnement juridique — une étape préalable de l'analyse juridique." }, "ja": { "name": "legal-element-extraction", "summary": "事件の記述、当事者の陳述、チャット記録、報道など非構造化テキストから法的意味を持つ事実を抽出する必要があるときに起動します。典型的な起動場面（これらに限りません）：日常語で書かれた事件記述から法律事実を抽出してほしい；口語的で感情的な語りを、法律分析に使える事実構造へ変換してほしい；事件の語りの中の「客観的事実」「主観的判断」「法律的評価」を区別してほしい；抽出した事実を特定の法律関係の構成要件に対応付けてほしい；事件のタイムライン整理や事実整理表の作成を求める；生活言語を法律言語に変換してほしい（例：「返済をいつまでも引き延ばしている」→「履行期限経過後の不履行」）。本スキルの核心的価値：雑然とした生活の語りを、法律推論に入れることのできる事実構造へ整理すること。法律分析の前置ステップです。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:legal-interpretation-argument", "slug": "skill-legal-interpretation-argument", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when the following situations arise during legal reasoning: (1) the meaning of a statutory text is ambiguous, with multiple possible readings; (2) applying a provision to concrete case facts is disputed; (3) it must be argued that a provision should/should not apply to a specific situation;" }, "human": { "zh-CN": { "name": "legal-interpretation-argument", "summary": "当法律推理过程中遇到以下情形时触发本技能： （1）法条文本含义模糊、存在多种理解可能； （2）法条适用于具体案件事实时产生争议； （3）需要论证某一法条应当/不应当适用于特定情形； （4）当事人双方对同一法条的理解存在分歧； （5）法条之间存在表面冲突，需要通过解释协调； （6）新型案件事实无法直接对应既有法条文本； （7）需要对裁判文书中的法律适用进行分析或质疑。 本技能的核心是运用文义解释、体系解释、目的解释等法律解释方法，对关键法条进行严谨的解释论证，得出合理、可辩护的法律含义，并以规范化的方式呈现解释过程与结论。" }, "en": { "name": "legal-interpretation-argument", "summary": "Triggers when the following situations arise during legal reasoning: (1) the meaning of a statutory text is ambiguous, with multiple possible readings; (2) applying a provision to concrete case facts is disputed; (3) it must be argued that a provision should/should not apply to a specific situation; (4) the two parties disagree on the understanding of the same provision; (5) provisions appear to conflict and need reconciliation through interpretation; (6) novel case facts cannot be directly mapped onto existing statutory texts; (7) the application of law in an adjudication document needs to be analyzed or challenged. The core of this skill is applying legal interpretation methods — literal, systemic, teleological, and others — to rigorously interpret and argue key provisions, reach reasonable and defensible legal meanings, and present the interpretive process and conclusions in a standardized way." }, "es": { "name": "legal-interpretation-argument", "summary": "Se activa cuando durante el razonamiento jurídico surgen las siguientes situaciones: (1) el significado del texto de un precepto es ambiguo y admite varias lecturas; (2) la aplicación de un precepto a los hechos concretos del caso es discutida; (3) hay que argumentar que un precepto debe/no debe aplicarse a una situación determinada; (4) ambas partes discrepan sobre el entendimiento del mismo precepto; (5) existen conflictos aparentes entre preceptos que requieren conciliación interpretativa; (6) hechos novedosos del caso no pueden corresponderse directamente con los textos legales existentes; (7) hay que analizar o cuestionar la aplicación del derecho en un documento judicial. El núcleo de esta habilidad consiste en emplear métodos de interpretación jurídica —literal, sistemática, teleológica, entre otros— para interpretar y argumentar rigurosamente los preceptos clave, llegar a significados jurídicos razonables y defendibles, y presentar el proceso y las conclusiones interpretativas de manera normalizada." }, "fr": { "name": "legal-interpretation-argument", "summary": "Se déclenche lorsque les situations suivantes apparaissent en raisonnement juridique : (1) le sens du texte d'une disposition est ambigu et admet plusieurs lectures ; (2) l'application d'une disposition aux faits concrets du dossier est contestée ; (3) il faut argumenter qu'une disposition doit/ne doit pas s'appliquer à une situation donnée ; (4) les deux parties divergent sur l'intelligence d'une même disposition ; (5) des dispositions sont en conflit apparent et doivent être conciliées par l'interprétation ; (6) des faits nouveaux du dossier ne peuvent être directement rattachés aux textes législatifs existants ; (7) l'application du droit dans un document juridictionnel doit être analysée ou contestée. Le cœur de cette compétence : employer les méthodes d'interprétation juridique — littérale, systémique, téléologique, etc. — pour interpréter et argumenter rigoureusement les dispositions clés, parvenir à des significations juridiques raisonnables et défendables, et présenter le processus et les conclusions d'interprétation de façon normalisée." }, "ja": { "name": "legal-interpretation-argument", "summary": "法律推論の過程で以下の状況に直面したときに起動します：(1) 条文テキストの意味が曖昧で複数の解釈の可能性がある；(2) 条文を具体的事件事実に適用することに争いがある；(3) ある条文が特定の場面に適用されるべきか否かを論証する必要がある；(4) 当事者間で同一条文の理解が分かれている；(5) 条文間に見かけ上の衝突があり、解釈による調整が必要；(6) 新類型の事件事実が既存の条文テキストに直接対応しない；(7) 裁判文書における法律適用の分析・批判が必要。本スキルの核心は、文理解釈、体系的解釈、目的論的解釈などの法律解釈手法を用いて、重要な条文を厳密に解釈・論証し、妥当で防御可能な法的意味を導き、解釈の過程と結論を規範的な形で提示することです。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:legal-judgment-prediction", "slug": "skill-legal-judgment-prediction", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Based on a description of case facts, comprehensively invokes 8 atomic capabilities to predict the charged offense, applicable provisions, sentence, and sentencing factors, outputting a structured prediction report with a confidence assessment." }, "human": { "zh-CN": { "name": "legal-judgment-prediction", "summary": "根据案件事实描述，综合调用8个原子能力，预测罪名、适用法条、刑期及量刑情节，输出结构化预测报告及置信度评估。" }, "en": { "name": "legal-judgment-prediction", "summary": "Based on a description of case facts, comprehensively invokes 8 atomic capabilities to predict the charged offense, applicable provisions, sentence, and sentencing factors, outputting a structured prediction report with a confidence assessment." }, "es": { "name": "legal-judgment-prediction", "summary": "A partir de la descripción de los hechos del caso, invoca de manera integral 8 capacidades atómicas para prever el tipo de delito imputado, los preceptos aplicables, la pena y las circunstancias de la punición, y produce un informe de previsión estructurado con una valoración de confianza." }, "fr": { "name": "legal-judgment-prediction", "summary": "À partir d'une description des faits du dossier, invoque de façon intégrée 8 capacités atomiques pour prédire le type d'infraction retenue, les dispositions applicables, la peine et les circonstances de la sanction, et produit un rapport de prédiction structuré avec une évaluation de confiance." }, "ja": { "name": "legal-judgment-prediction", "summary": "事件事実の記述に基づき、8つのアトミック能力を総合的に呼び出して、罪名、適用法条、刑期、量刑情状を予測し、構造化された予測レポートと信頼度評価を出力します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:legal-norm-validity-check", "slug": "skill-legal-norm-validity-check", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when the AI agent, having retrieved specific legal provisions during reasoning, needs to verify their validity. Trigger conditions include but are not limited to: 1. Before the agent cites any legal provision as a basis for reasoning; 2. When the user provides a provision and asks for an ap" }, "human": { "zh-CN": { "name": "legal-norm-validity-check", "summary": "当AI智能体在法律推理过程中检索到具体法律条文后，需要对该法条进行效力验证时触发此技能。 触发条件包括但不限于： 1. 智能体引用任何法律条文作为推理依据前； 2. 用户提供法条要求分析其适用性时； 3. 多个法律规范可能存在冲突需要判断优先适用时； 4. 法律文件的时间效力存疑（可能已被修订、废止）时； 5. 需要确认某规范性文件是否具有法律约束力时； 6. 跨层级、跨区域法律规范适用产生疑问时。 本技能确保智能体引用的每一条法律规范均为现行有效、层级正确、与上位法及同位法无冲突， 从而保障法律推理结论的可靠性和权威性。" }, "en": { "name": "legal-norm-validity-check", "summary": "Triggers when the AI agent, having retrieved specific legal provisions during reasoning, needs to verify their validity. Trigger conditions include but are not limited to: 1. Before the agent cites any legal provision as a basis for reasoning; 2. When the user provides a provision and asks for an applicability analysis; 3. When multiple legal norms may conflict and the priority of application must be judged; 4. When a legal document's temporal validity is in doubt (it may have been amended or repealed); 5. When it must be confirmed whether a normative document is legally binding; 6. When applying legal norms across levels or regions raises questions. This skill ensures every legal norm the agent cites is currently effective, at the correct hierarchy level, and conflict-free with higher-level and same-level norms, thereby safeguarding the reliability and authority of the reasoning conclusions." }, "es": { "name": "legal-norm-validity-check", "summary": "Se activa cuando el agente de IA, tras recuperar preceptos legales concretos durante el razonamiento, necesita verificar su validez. Condiciones de activación incluyen, entre otras: 1. Antes de que el agente cite cualquier precepto legal como base de razonamiento; 2. Cuando el usuario aporta un precepto y solicita analizar su aplicabilidad; 3. Cuando varias normas jurídicas pueden entrar en conflicto y hay que juzgar la prelación de aplicación; 4. Cuando la vigencia temporal de un documento legal es dudosa (posiblemente modificado o derogado); 5. Cuando hay que confirmar si un documento normativo tiene fuerza vinculante; 6. Cuando la aplicación de normas jurídicas entre niveles o regiones plantea dudas. Esta habilidad garantiza que cada norma jurídica citada por el agente esté vigente, en el nivel jerárquico correcto y sin conflicto con normas superiores o del mismo nivel, protegiendo así la fiabilidad y la autoridad de las conclusiones del razonamiento." }, "fr": { "name": "legal-norm-validity-check", "summary": "Se déclenche lorsque l'agent IA, après avoir retrouvé des dispositions légales précises en cours de raisonnement, doit vérifier leur validité. Conditions de déclenchement, sans s'y limiter : 1. Avant que l'agent ne cite une disposition légale comme fondement du raisonnement ; 2. Lorsque l'utilisateur fournit une disposition et demande l'analyse de son applicabilité ; 3. Lorsque plusieurs normes juridiques peuvent entrer en conflit et qu'il faut juger la primauté d'application ; 4. Lorsque la validité temporelle d'un document juridique est douteuse (possiblement modifié ou abrogé) ; 5. Lorsqu'il faut confirmer qu'un document normatif est juridiquement contraignant ; 6. Lorsque l'application de normes juridiques entre niveaux ou régions soulève des doutes. Cette compétence garantit que chaque norme juridique citée par l'agent est en vigueur, au bon niveau hiérarchique et sans conflit avec les normes supérieures ou de même niveau, protégeant ainsi la fiabilité et l'autorité des conclusions du raisonnement." }, "ja": { "name": "legal-norm-validity-check", "summary": "AIエージェントが法律推論の過程で具体的な法律条文を検索した後、その条文の効力検証が必要なときに起動します。起動条件（これらに限りません）：1. エージェントがいかなる法律条文を推論の根拠として引用する前；2. ユーザーが条文を提示し、その適用可能性の分析を求めるとき；3. 複数の法律規範が衝突する可能性があり、優先適用を判断する必要があるとき；4. 法律文書の時間的効力に疑問がある（改正・廃止された可能性がある）とき；5. ある規範的文書が法的拘束力を持つか確認が必要なとき；6. 階層や地域をまたぐ法律規範の適用に疑問が生じたとき。本スキルは、エージェントが引用するすべての法律規範が現行有効であり、階層が正しく、上位法・同位法と衝突しないことを保証し、法律推論結論の信頼性と権威性を保障します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:legal-risk-assessment", "slug": "skill-legal-risk-assessment", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Evaluates the regulatory penalty risk facing an enterprise along the four dimensions of licensing qualifications, regulatory compliance, and historical penalty records." }, "human": { "zh-CN": { "name": "legal-risk-assessment", "summary": "从许可资质、监管法规遵循及历史处罚记录四个维度评估企业面临的监管处罚风险。" }, "en": { "name": "legal-risk-assessment", "summary": "Evaluates the regulatory penalty risk facing an enterprise along the four dimensions of licensing qualifications, regulatory compliance, and historical penalty records." }, "es": { "name": "legal-risk-assessment", "summary": "Evalúa el riesgo de sanción regulatoria al que se enfrenta una empresa a lo largo de las cuatro dimensiones de licencias y calificaciones, cumplimiento normativo y antecedentes de sanciones." }, "fr": { "name": "legal-risk-assessment", "summary": "Évalue le risque de sanction réglementaire auquel une entreprise est exposée selon les quatre dimensions que sont les licences et qualifications, la conformité réglementaire et les antécédents de sanctions." }, "ja": { "name": "legal-risk-assessment", "summary": "許認可資格、規制法令の遵守、過去の処分記録という4つの次元から、企業が直面する規制裁罰リスクを評価します。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:legal-terminology", "slug": "skill-legal-terminology", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "When the AI agent generates, reviews, or modifies any legal text (including but not limited to contracts, pleadings, legal opinions, adjudication documents, statute annotations, and legal memoranda), it must ensure that the legal terminology used is worded accurately, without ambiguity, and in line " }, "human": { "zh-CN": { "name": "legal-terminology", "summary": "当AI智能体在生成、审查或修改任何法律文本（包括但不限于合同、诉状、法律意见书、裁判文书、法规释义、法律备忘录等）时，需要确保其中使用的法律术语措辞准确、无歧义、符合法律文体要求。该技能在以下场景触发： 1. 起草或修改法律文书时，需要选择正确的法律术语； 2. 审查法律文本时，发现术语使用不规范、混淆或存在歧义； 3. 将日常用语转化为法律规范表达时； 4. 翻译或对照中外法律术语时； 5. 用户明确要求对法律文本进行术语规范化处理时； 6. 对法律概念进行精确定义或解释时。 本技能是法律写作的基础原子能力，贯穿所有法律文本生产环节。" }, "en": { "name": "legal-terminology", "summary": "When the AI agent generates, reviews, or modifies any legal text (including but not limited to contracts, pleadings, legal opinions, adjudication documents, statute annotations, and legal memoranda), it must ensure that the legal terminology used is worded accurately, without ambiguity, and in line with legal style requirements. This skill triggers in the following scenarios: 1. selecting the correct legal terms when drafting or revising legal documents; 2. detecting non-standard, confused, or ambiguous terminology when reviewing legal texts; 3. converting everyday language into normative legal expression; 4. translating or aligning Chinese and foreign legal terms; 5. the user explicitly requests terminology normalization of a legal text; 6. precisely defining or explaining a legal concept. This skill is the foundational atomic capability of legal writing and runs through all stages of legal text production." }, "es": { "name": "legal-terminology", "summary": "Cuando el agente de IA genera, revisa o modifica cualquier texto jurídico (incluyendo, entre otros, contratos, escritos de demanda, dictámenes jurídicos, documentos judiciales, comentarios de normas y memorandos jurídicos), debe garantizar que la terminología jurídica empleada esté redactada con precisión, sin ambigüedad y conforme a los requisitos del estilo jurídico. Esta habilidad se activa en los siguientes escenarios: 1. seleccionar los términos jurídicos correctos al redactar o revisar documentos jurídicos; 2. detectar terminología impropia, confusa o ambigua al revisar textos jurídicos; 3. convertir lenguaje cotidiano en expresión jurídica normativa; 4. traducir o cotejar términos jurídicos chinos y extranjeros; 5. el usuario solicita expresamente la normalización terminológica de un texto jurídico; 6. definir o explicar con precisión un concepto jurídico. Esta habilidad es la capacidad atómica básica de la redacción jurídica y atraviesa todas las fases de producción de textos jurídicos." }, "fr": { "name": "legal-terminology", "summary": "Lorsque l'agent IA génère, révise ou modifie tout texte juridique (y compris, sans s'y limiter, contrats, conclusions, avis juridiques, documents judiciaires, commentaires de textes et mémos juridiques), il doit garantir que la terminologie juridique employée est formulée avec exactitude, sans ambiguïté et conforme aux exigences du style juridique. Cette compétence se déclenche dans les scénarios suivants : 1. choisir les termes juridiques corrects lors de la rédaction ou de la révision de documents juridiques ; 2. repérer une terminologie impropre, confuse ou ambiguë lors de la revue de textes juridiques ; 3. convertir le langage courant en expression juridique normative ; 4. traduire ou confronter des termes juridiques chinois et étrangers ; 5. l'utilisateur demande explicitement la normalisation terminologique d'un texte juridique ; 6. définir ou expliquer avec précision un concept juridique. Cette compétence est la capacité atomique de base de la rédaction juridique et traverse toutes les étapes de la production de textes juridiques." }, "ja": { "name": "legal-terminology", "summary": "AIエージェントがあらゆる法律テキスト（契約書、訴状、法律意見書、裁判文書、法令解釈、法律メモなど。これらに限りません）を生成・審査・修正する際には、用いられる法律用語が正確に表現され、曖昧さがなく、法律文体の要求に適合することを確保する必要があります。このスキルは次の場面で起動します：1. 法律文書の起草・修正時に正しい法律用語を選ぶ必要がある；2. 法律テキストの審査時に、用語の不適切な使用、混同、曖昧さを発見した；3. 日常語を法律規範的表現に変換する；4. 中国語と外国語の法律用語を翻訳・対照する；5. ユーザーが法律テキストの用語の標準化を明示的に求める；6. 法律概念を厳密に定義または説明する。本スキルは法律執筆の基礎的なアトミック能力であり、法律テキスト生産の全工程を貫きます。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:multi-document-summarization", "slug": "skill-multi-document-summarization", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when multiple related documents need comprehensive analysis, extraction of common viewpoints, identification of conflicts, and generation of a unified overview. Typical trigger scenarios include but are not limited to: the user uploads multiple judgments or rulings and asks to summarize the" }, "human": { "zh-CN": { "name": "multi-document-summarization", "summary": "当需要对多份相关文档进行综合分析、提取共同观点、识别冲突并生成统一概览时触发此技能。 典型触发场景包括但不限于： - 用户上传多份判决书、裁定书，要求总结类案裁判要旨 - 用户提供多个合同文本，要求归纳条款异同和关键差异 - 用户提供多份法律法规、司法解释，要求梳理规范演变和适用要点 - 用户提供多个证据材料、调查笔录，要求整合事实脉络 - 用户提供多篇学术论文、研究报告，要求提炼研究共识和分歧 - 用户要求对比多份法律意见书的观点差异 本技能的核心价值：不是简单压缩单篇文档，而是构建跨文档的关联分析，识别共性与差异，生成新的综合性洞见。" }, "en": { "name": "multi-document-summarization", "summary": "Triggers when multiple related documents need comprehensive analysis, extraction of common viewpoints, identification of conflicts, and generation of a unified overview. Typical trigger scenarios include but are not limited to: the user uploads multiple judgments or rulings and asks to summarize the gist of rulings in similar cases; provides multiple contract texts and asks to summarize clause similarities and key differences; provides multiple laws, regulations, or judicial interpretations and asks to trace normative evolution and application points; provides multiple evidence materials or investigation transcripts and asks to integrate the factual thread; provides multiple academic papers or research reports and asks to distill research consensus and disagreements; asks to compare viewpoint differences across multiple legal opinions. Core value of this skill: not simply compressing a single document, but building cross-document relational analysis, identifying commonalities and differences, and generating new synthesized insights." }, "es": { "name": "multi-document-summarization", "summary": "Se activa cuando varios documentos relacionados requieren análisis integral, extracción de puntos de vista comunes, identificación de conflictos y generación de una vista unificada. Escenarios típicos de activación incluyen, entre otros: el usuario sube múltiples sentencias o providencias y pide resumir la doctrina esencial de los casos análogos; aporta varios textos contractuales y pide resumir semejanzas y diferencias clave entre cláusulas; aporta múltiples leyes, reglamentos o interpretaciones judiciales y pide trazar la evolución normativa y los puntos de aplicación; aporta múltiples materiales probatorios o actas de investigación y pide integrar el hilo fáctico; aporta varios artículos académicos o informes de investigación y pide destilar consensos y discrepancias de investigación; pide comparar las diferencias de criterio entre varios dictámenes jurídicos. Valor central de esta habilidad: no comprimir simplemente un documento aislado, sino construir un análisis relacional entre documentos, identificar coincidencias y diferencias, y generar nuevas percepciones sintéticas." }, "fr": { "name": "multi-document-summarization", "summary": "Se déclenche lorsque plusieurs documents apparentés exigent une analyse intégrée, l'extraction des points de vue communs, l'identification des conflits et la production d'une vue unifiée. Scénarios types de déclenchement, sans s'y limiter : l'utilisateur téléverse plusieurs jugements ou ordonnances et demande de résumer les motifs essentiels des affaires analogues ; fournit plusieurs textes contractuels et demande de résumer similitudes et différences clés entre clauses ; fournit plusieurs lois, règlements ou interprétations judiciaires et demande de retracer l'évolution normative et les points d'application ; fournit plusieurs pièces probatoires ou procès-verbaux d'enquête et demande d'intégrer le fil factuel ; fournit plusieurs articles académiques ou rapports de recherche et demande de dégager les consensus et divergences de recherche ; demande de comparer les divergences de vues entre plusieurs avis juridiques. Valeur centrale de cette compétence : non pas compresser un simple document, mais construire une analyse relationnelle transversale, identifier convergences et différences, et produire de nouvelles synthèses éclairantes." }, "ja": { "name": "multi-document-summarization", "summary": "複数の関連文書を統合分析し、共通の見解を抽出し、衝突を識別し、統一の概観を生成する必要があるときに起動します。典型的な起動場面（これらに限りません）：複数の判決書・裁定書をアップロードし、類似事件の裁判要旨の要約を求める；複数の契約書テキストを提供し、条項の異同と重要な差異を帰納してほしい；複数の法令・司法解釈を提供し、規範の変遷と適用の要点を整理してほしい；複数の証拠資料・調査筆録を提供し、事実の筋道を統合してほしい；複数の学術論文・研究報告を提供し、研究上の合意と相違を抽出してほしい；複数の法律意見書の見解の違いを比較してほしい。本スキルの核心的価値：単一文書の単純圧縮ではなく、文書横断の関連分析を構築し、共通点と相違点を識別し、新たな総合的洞察を生み出すことです。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:other-legal-retrieval", "slug": "skill-other-legal-retrieval", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Triggers when the agent needs to retrieve auxiliary legal information beyond statutes, judicial interpretations, and typical cases. Specifically includes but is not limited to: legislative background materials (draft explanations, deliberation reports, legislators' press Q&A), regulatory enforcement" }, "human": { "zh-CN": { "name": "other-legal-retrieval", "summary": "当智能体需要检索法律条文、司法解释、典型案例之外的辅助性法律信息时触发此技能。具体包括但不限于：立法背景资料（草案说明、审议报告、立法机关答记者问）、监管案例与行政处罚决定、地方指导意见与会议纪要、部门规章的起草说明与政策解读、行业标准与技术规范、法律实施情况报告、人大常委会执法检查报告、白皮书与政策文件、学术通说与权威教材观点、域外法律比较资料等。当用户问题涉及'为什么这样规定''立法目的是什么''实践中监管部门怎么处理''地方上有什么特殊规定''行业惯例是什么'等语境时，应当触发本技能。" }, "en": { "name": "other-legal-retrieval", "summary": "Triggers when the agent needs to retrieve auxiliary legal information beyond statutes, judicial interpretations, and typical cases. Specifically includes but is not limited to: legislative background materials (draft explanations, deliberation reports, legislators' press Q&A), regulatory enforcement cases and administrative penalty decisions, local guidance documents and meeting minutes, drafting explanations and policy readings of departmental rules, industry standards and technical specifications, law-implementation status reports, NPC standing-committee law-enforcement inspection reports, white papers and policy documents, academic prevailing views and authoritative textbook positions, comparative foreign-law materials, and more. This skill should be triggered when the user's question involves contexts such as \"why is it regulated this way\", \"what is the legislative purpose\", \"how do regulators handle it in practice\", \"what special local provisions exist\", or \"what are the industry conventions\"." }, "es": { "name": "other-legal-retrieval", "summary": "Se activa cuando el agente necesita recuperar información jurídica auxiliar más allá de los preceptos legales, las interpretaciones judiciales y los casos típicos. Incluye específicamente, entre otros: materiales de contexto legislativo (exposiciones de anteproyectos, informes de deliberación, respuestas de los legisladores a la prensa), casos de supervisión y resoluciones sancionadoras administrativas, directrices locales y actas de reuniones, exposiciones de elaboración y lecturas de política de los reglamentos departamentales, normas del sector y especificaciones técnicas, informes de estado de aplicación de la ley, informes de inspección de aplicación de la ley del comité permanente de la Asamblea Popular, libros blancos y documentos de política, doctrina académica mayoritaria y posiciones de manuales que hacen autoridad, materiales de derecho comparado extranjero, etc. Esta habilidad debe activarse cuando la pregunta del usuario entraña contextos como \"¿por qué se regula así?\", \"¿cuál es el propósito legislativo?\", \"¿cómo lo manejan los órganos de supervisión en la práctica?\", \"¿qué disposiciones locales especiales hay?\" o \"¿cuáles son los usos del sector?\"." }, "fr": { "name": "other-legal-retrieval", "summary": "Se déclenche lorsque l'agent doit retrouver des informations juridiques auxiliaires au-delà des textes de loi, des interprétations judiciaires et des affaires types. Comprend spécifiquement, sans s'y limiter : les documents d'arrière-plan législatif (exposés des projets, rapports de délibération, réponses des législateurs à la presse), les affaires des régulateurs et les décisions de sanction administrative, les directives locales et comptes rendus de réunions, les exposés d'élaboration et lectures de politique des règlements ministériels, les normes sectorielles et spécifications techniques, les rapports de situation d'application de la loi, les rapports d'inspection de l'application des lois du comité permanent de l'Assemblée populaire, les livres blancs et documents de politique publique, les thèses académiques dominantes et positions de manuels faisant autorité, les matériaux de droit comparé étranger, etc. Cette compétence doit se déclencher lorsque la question de l'utilisateur relève de contextes tels que « pourquoi cette règle ? », « quel est le but législatif ? », « comment les autorités de régulation procèdent-elles en pratique ? », « quelles dispositions locales particulières existent ? » ou « quelles sont les pratiques du secteur ? »." }, "ja": { "name": "other-legal-retrieval", "summary": "エージェントが法律条文、司法解釈、典型事件以外の補助的な法律情報を検索する必要があるときに起動します。具体的には（これらに限りません）：立法背景資料（草案説明、審議報告、立法機関の記者会見応答）、規制事例と行政処分決定、地方の指導意見と会議要録、部門規則の起草説明と政策解読、業界標準と技術規範、法律実施状況報告、人大常務委員会の法執行検査報告、白書と政策文書、学術通説と権威ある教科書の見解、域外法律比較資料など。ユーザーの質問が「なぜこのように規定されているのか」「立法目的は何か」「実務で監督管理部門はどう処理するか」「地方にはどんな特別規定があるか」「業界の慣行は何か」といった文脈を含むとき、本スキルを起動すべきです。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:strategic-risk-prioritization", "slug": "skill-strategic-risk-prioritization", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "When legal analysis contains multiple possible conclusions, risk points, or disputed issues, these risks need to be systematically ranked by probability of occurrence and degree of impact, helping decision-makers focus on the most critical risks and allocate resources sensibly. This skill triggers i" }, "human": { "zh-CN": { "name": "strategic-risk-prioritization", "summary": "当法律分析中存在多个可能的结论、风险点或争议焦点时，需要对这些风险按照发生概率和影响程度进行系统化排序，以帮助决策者聚焦最关键的风险并合理分配资源。该技能在以下场景触发：(1)法律尽职调查中发现多项合规风险需要排序；(2)诉讼策略制定中需要评估多个不利结论的可能性与后果；(3)交易结构设计中需要识别并排列关键法律障碍；(4)合同审查中发现多个风险条款需要确定修改优先级；(5)监管合规评估中需要对多项违规风险进行分级管理；(6)任何需要在有限资源下对法律风险进行战略性取舍的场景。" }, "en": { "name": "strategic-risk-prioritization", "summary": "When legal analysis contains multiple possible conclusions, risk points, or disputed issues, these risks need to be systematically ranked by probability of occurrence and degree of impact, helping decision-makers focus on the most critical risks and allocate resources sensibly. This skill triggers in the following scenarios: (1) multiple compliance risks found in legal due diligence that need ranking; (2) litigation strategy design requiring assessment of the likelihood and consequences of multiple adverse conclusions; (3) transaction structure design requiring identification and ordering of key legal obstacles; (4) multiple risky clauses found in contract review requiring determination of modification priorities; (5) regulatory compliance assessment requiring tiered management of multiple violation risks; (6) any scenario requiring strategic trade-offs among legal risks under limited resources." }, "es": { "name": "strategic-risk-prioritization", "summary": "Cuando el análisis jurídico contiene múltiples conclusiones posibles, puntos de riesgo o puntos de disputa, estos riesgos deben ordenarse sistemáticamente por probabilidad de ocurrencia y grado de impacto, para ayudar a los decisores a centrarse en los riesgos más críticos y asignar los recursos de manera razonable. Esta habilidad se activa en los siguientes escenarios: (1) múltiples riesgos de cumplimiento hallados en la due diligence jurídica que requieren ordenación; (2) diseño de la estrategia litigiosa que exige evaluar la probabilidad y las consecuencias de varias conclusiones desfavorables; (3) diseño de la estructura de la transacción que exige identificar y ordenar los obstáculos jurídicos clave; (4) múltiples cláusulas de riesgo halladas en la revisión contractual que exigen fijar prioridades de modificación; (5) evaluación de cumplimiento regulatorio que exige gestión escalonada de múltiples riesgos de infracción; (6) cualquier escenario que exija decisiones estratégicas entre riesgos jurídicos con recursos limitados." }, "fr": { "name": "strategic-risk-prioritization", "summary": "Lorsque l'analyse juridique contient plusieurs conclusions possibles, points de risque ou points litigieux, ces risques doivent être classés systématiquement selon leur probabilité d'occurrence et leur degré d'impact, pour aider les décideurs à se concentrer sur les risques les plus critiques et à allouer raisonnablement les ressources. Cette compétence se déclenche dans les scénarios suivants : (1) plusieurs risques de conformité identifiés en due diligence juridique à classer ; (2) conception de la stratégie contentieuse exigeant d'évaluer la probabilité et les conséquences de plusieurs conclusions défavorables ; (3) conception du montage transactionnel exigeant d'identifier et d'ordonner les obstacles juridiques clés ; (4) plusieurs clauses à risque identifiées lors de la revue de contrats exigeant de prioriser les modifications ; (5) évaluation de conformité réglementaire exigeant une gestion hiérarchisée de multiples risques de violation ; (6) tout scénario exigeant des arbitrages stratégiques entre risques juridiques sous ressources limitées." }, "ja": { "name": "strategic-risk-prioritization", "summary": "法律分析において複数の可能な結論、リスクポイント、争点が存在する場合、発生確率と影響度に基づいてこれらのリスクを体系的に順位付けし、意思決定者が最も重要なリスクに集中し資源を合理的に配分できるようにする必要があります。このスキルは次の場面で起動します：(1) 法的デューデリジェンスで複数のコンプライアンスリスクが見つかり順位付けが必要；(2) 訴訟戦略の立案で複数の不利益結論の可能性と帰結の評価が必要；(3) 取引ストラクチャー設計で重要な法的障害の特定と順序付けが必要；(4) 契約審査で複数のリスク条項が見つかり修正の優先順位を確定する必要がある；(5) 規制コンプライアンス評価で複数の違反リスクをレベル別に管理する必要がある；(6) 限られた資源の下で法的リスクを戦略的に取捨選択する必要があるあらゆる場面。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:structured-element-extraction", "slug": "skill-structured-element-extraction", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "When the AI agent needs to systematically analyze a legal issue, case facts, a legal provision, or a legal relation, it must first decompose it into a structured checklist of elements. This skill triggers in the following scenarios: receiving a description of case facts that requires extracting all " }, "human": { "zh-CN": { "name": "structured-element-extraction", "summary": "当AI智能体需要对一个法律问题、案件事实、法律条文或法律关系进行系统性分析时，首先必须将其分解为结构化的要素清单。该技能在以下场景触发： - 接收到一段案件事实描述，需要提取全部法律要素后再进行分析 - 需要判断某一法律关系是否成立（如合同成立、侵权构成、犯罪构成） - 需要对比两个或多个法律事实/条文的异同 - 在进入法律推理、法律适用、法律建议等下游任务之前，作为前置质量关卡 - 用户明确要求\"列出要素\"\"逐项分析\"\"检查是否遗漏\" 本技能的核心目标是：确保要素提取完整、无遗漏，形成可追溯、可验证的结构化清单，然后才允许进入下一阶段推理。它是法律分析链条中的\"质量闸门\"。" }, "en": { "name": "structured-element-extraction", "summary": "When the AI agent needs to systematically analyze a legal issue, case facts, a legal provision, or a legal relation, it must first decompose it into a structured checklist of elements. This skill triggers in the following scenarios: receiving a description of case facts that requires extracting all legal elements before analysis; determining whether a legal relation is established (e.g., formation of a contract, the elements of a tort, the elements of a crime); comparing two or more legal facts/provisions for similarities and differences; serving as a preliminary quality gate before downstream tasks such as legal reasoning, application of law, and legal advice; the user explicitly asks to \"list the elements\", \"analyze item by item\", or \"check for omissions\". The core goal of this skill: ensure element extraction is complete and without omissions, producing a traceable, verifiable structured checklist before the next reasoning stage is allowed. It is the \"quality gate\" in the chain of legal analysis." }, "es": { "name": "structured-element-extraction", "summary": "Cuando el agente de IA necesita analizar sistemáticamente un problema jurídico, hechos del caso, un precepto legal o una relación jurídica, primero debe descomponerlo en una lista estructurada de elementos. Esta habilidad se activa en los siguientes escenarios: recibir una descripción de hechos del caso que exige extraer todos los elementos jurídicos antes de analizar; determinar si una relación jurídica se constituye (p. ej., formación del contrato, elementos de la responsabilidad extracontractual, elementos del delito); comparar dos o más hechos/preceptos jurídicos en semejanzas y diferencias; actuar como control de calidad previo antes de tareas aguas abajo como el razonamiento jurídico, la aplicación del derecho y el asesoramiento jurídico; el usuario pide explícitamente \"listar los elementos\", \"analizar punto por punto\" o \"comprobar si falta algo\". Objetivo central de esta habilidad: garantizar una extracción completa y sin omisiones, produciendo una lista estructurada rastreable y verificable antes de permitir la siguiente fase de razonamiento. Es la \"puerta de calidad\" de la cadena de análisis jurídico." }, "fr": { "name": "structured-element-extraction", "summary": "Lorsque l'agent IA doit analyser systématiquement une question juridique, des faits de dossier, une disposition légale ou une relation juridique, il doit d'abord la décomposer en une liste structurée d'éléments. Cette compétence se déclenche dans les scénarios suivants : réception d'une description de faits de dossier exigeant l'extraction de tous les éléments juridiques avant analyse ; détermination de la constitution d'une relation juridique (ex. : formation du contrat, éléments de la responsabilité délictuelle, éléments de l'infraction) ; comparaison de deux faits/dispositions juridiques ou plus ; rôle de porte qualité préalable avant les tâches aval telles que le raisonnement juridique, l'application du droit et les conseils juridiques ; l'utilisateur demande explicitement de « lister les éléments », « analyser point par point » ou « vérifier les omissions ». Objectif central de cette compétence : garantir une extraction complète et sans omission, produire une liste structurée traçable et vérifiable avant d'autoriser la phase de raisonnement suivante. C'est la « porte de qualité » de la chaîne d'analyse juridique." }, "ja": { "name": "structured-element-extraction", "summary": "AIエージェントが法律問題、事件事実、法律条文、法律関係を体系的に分析する必要があるときは、まずそれを構造化された要素リストに分解しなければなりません。このスキルは次の場面で起動します：事件事実の記述を受け取り、分析前にすべての法律要素を抽出する必要がある；ある法律関係が成立するか判断する必要がある（契約の成立、不法行為の構成、犯罪の構成など）；2つ以上の法律事実・条文の異同を比較する必要がある；法律推論、法律適用、法律アドバイスなどの下流タスクに入る前の前置きの品質ゲートとして；ユーザーが「要素を列挙」「一項ずつ分析」「抜け漏れチェック」を明示的に求める。本スキルの核心目標：要素抽出の完全性と抜けのなさを確保し、追跡可能で検証可能な構造化リストを作成した上で、次段階の推論に入ることを許可すること。法律分析連鎖における「品質ゲート」です。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:systematic-interpretation", "slug": "skill-systematic-interpretation", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Systematic interpretation addresses disputes over meaning or scope of application that arise after the literal interpretation of a legal norm, producing the interpretation most consistent with the system — based on the norm's position within the legal system (which law, part, chapter, or section it " }, "human": { "zh-CN": { "name": "systematic-interpretation", "summary": "体系解释是指针对法律规范文义解释后所产生的含义或适用范围上的争议，根据法律规范在法律体系中的位置（如在哪一部法律、编、章或节），关联的法律规范的含义，或者法律作为一个体系的应由之意，做出最符合体系性的解释，从而明确体系视角下法律规范的含义与适用范围。 触发条件（满足任一即可调用）：" }, "en": { "name": "systematic-interpretation", "summary": "Systematic interpretation addresses disputes over meaning or scope of application that arise after the literal interpretation of a legal norm, producing the interpretation most consistent with the system — based on the norm's position within the legal system (which law, part, chapter, or section it belongs to), the meanings of the related legal norms, or the sense the law must bear as a system — thereby clarifying the meaning and scope of application of the norm from a systemic perspective. Trigger conditions (any one suffices to invoke):" }, "es": { "name": "systematic-interpretation", "summary": "La interpretación sistemática aborda las controversias sobre el significado o el alcance de aplicación surgidas tras la interpretación literal de una norma jurídica, y produce la interpretación más coherente con el sistema —a partir de la posición de la norma en el ordenamiento jurídico (a qué ley, parte, capítulo o sección pertenece), del significado de las normas jurídicas vinculadas, o del sentido que el derecho debe tener como sistema—, aclarando así el significado y el alcance de aplicación de la norma desde la perspectiva sistémica. Condiciones de activación (basta con que se cumpla una para invocarla):" }, "fr": { "name": "systematic-interpretation", "summary": "L'interprétation systémique traite les contestations de sens ou de champ d'application nées après l'interprétation littérale d'une norme juridique, et produit l'interprétation la plus conforme au système — à partir de la place de la norme dans le système juridique (à quelle loi, partie, chapitre ou section elle appartient), du sens des normes juridiques connexes, ou du sens que le droit doit porter en tant que système — clarifiant ainsi le sens et le champ d'application de la norme du point de vue systémique. Conditions de déclenchement (une seule suffit pour l'invoquer) :" }, "ja": { "name": "systematic-interpretation", "summary": "体系的解釈とは、法律規範の文理解釈の後に生じる意味や適用範囲の争いに対して、法律体系の中での規範の位置（どの法律・編・章・節に属するか）、関連する法律規範の意味、あるいは法律が体系として担うべき意味に基づいて、最も体系性に合致する解釈を行い、体系の視点から規範の意味と適用範囲を明確にすることです。起動条件（いずれか一つを満たせば呼び出し可）：" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:teleological-interpretation", "slug": "skill-teleological-interpretation", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Teleological interpretation skill. When literal interpretation of a legal provision still yields no definite conclusion, it systematically discovers and argues the provision's purpose and, within what the text can bear, selects the most justified meaning to form an adjudicable interpretive conclusio" }, "human": { "zh-CN": { "name": "teleological-interpretation", "summary": "目的解释技能。当法律条文经过文义解释后仍无法得出确定结论时，通过系统性地发现并论证该条文的目的，在文本可承受范围内选择最正当的含义，形成可裁判的解释结论。触发条件（满足任一即可）：文义解释存在两种以上可能无法择一；文义解释结论明显不合理；条文含开放性评价概念（如\"合理\"\"正当\"\"必要\"\"重大\"）；条文较旧、面对新型事实时文义覆盖不足。" }, "en": { "name": "teleological-interpretation", "summary": "Teleological interpretation skill. When literal interpretation of a legal provision still yields no definite conclusion, it systematically discovers and argues the provision's purpose and, within what the text can bear, selects the most justified meaning to form an adjudicable interpretive conclusion. Trigger conditions (any one suffices): literal interpretation has two or more possible readings with no way to choose between them; the result of the literal interpretation is clearly unreasonable; the provision contains open-ended evaluative concepts (e.g., \"reasonable\", \"justified\", \"necessary\", \"significant\"); the provision is older and its literal coverage falls short against novel facts." }, "es": { "name": "teleological-interpretation", "summary": "Habilidad de interpretación teleológica. Cuando la interpretación literal de un precepto legal sigue sin permitir llegar a una conclusión determinada, descubre y argumenta sistemáticamente el propósito del precepto y, dentro de lo que el texto puede soportar, elige el significado más justificado para formar una conclusión interpretativa adjudicable. Condiciones de activación (basta una): la interpretación literal ofrece dos o más lecturas sin poder elegir entre ellas; el resultado de la interpretación literal es manifiestamente irrazonable; el precepto contiene conceptos valorativos abiertos (p. ej., \"razonable\", \"justo\", \"necesario\", \"grave\"); el precepto es antiguo y su cobertura literal resulta insuficiente ante hechos novedosos." }, "fr": { "name": "teleological-interpretation", "summary": "Compétence d'interprétation téléologique. Lorsque l'interprétation littérale d'une disposition juridique ne permet toujours pas d'aboutir à une conclusion déterminée, elle découvre et argumente systématiquement la finalité de la disposition et, dans les limites supportables du texte, choisit la signification la plus justifiée pour former une conclusion interprétative jugeable. Conditions de déclenchement (une seule suffit) : l'interprétation littérale offre deux lectures ou plus impossibles à départager ; le résultat de l'interprétation littérale est manifestement déraisonnable ; la disposition contient des concepts évaluatifs ouverts (ex. : « raisonnable », « juste », « nécessaire », « important ») ; la disposition est ancienne et sa couverture littérale est insuffisante face à des faits inédits." }, "ja": { "name": "teleological-interpretation", "summary": "目的論的解釈のスキルです。法律条文の文理解釈によっても確定的な結論が得られないとき、条文の目的を体系的に発見・論証し、テキストが許容する範囲内で最も正当な意味を選択して、裁判に供する解釈結論を形成します。起動条件（いずれか一つを満たせばよい）：文理解釈に2つ以上の可能性があり択一できない；文理解釈の結論が明らかに不合理である；条文に開放的な評価概念（「合理的」「正当」「必要」「重大」など）が含まれる；条文が古く、新類型の事実に対して文義の射程が不足している。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:trial-scheduling-and-deadline-monitoring", "slug": "skill-trial-scheduling-and-deadline-monitoring", "category": "skill", "marketCategory": null, "tags": [ "legal", "chinese-law", "thuyran" ], "canonical": { "brief": "Use this skill when the user needs to track, or be reminded of, the various statutory deadlines of a case — trial scheduling, enforcement scheduling, evidence submission deadlines, appeal deadlines, service deadlines, preservation renewals, and other statutory time limits. Typical trigger scenarios " }, "human": { "zh-CN": { "name": "trial-scheduling-and-deadline-monitoring", "summary": "当用户需要跟踪、被提醒案件的开庭排期、执行排期、证据提交期限、上诉期限、送达期限、保全续保等各项法定期限时，使用此技能。 典型触发场景包括但不限于： - 提到\"上诉期限\"\"证据提交期限\"\"送达期限\"\"保全期限\"等关键词 - 增加、修改、询问案件期限时" }, "en": { "name": "trial-scheduling-and-deadline-monitoring", "summary": "Use this skill when the user needs to track, or be reminded of, the various statutory deadlines of a case — trial scheduling, enforcement scheduling, evidence submission deadlines, appeal deadlines, service deadlines, preservation renewals, and other statutory time limits. Typical trigger scenarios include but are not limited to: mentioning keywords such as \"appeal deadline\", \"evidence submission deadline\", \"service deadline\", or \"preservation deadline\"; adding, modifying, or asking about case deadlines." }, "es": { "name": "trial-scheduling-and-deadline-monitoring", "summary": "Use esta habilidad cuando el usuario necesite hacer seguimiento de los distintos plazos legales de un caso —agenda de audiencias, agenda de ejecución, plazos de presentación de pruebas, plazos de apelación, plazos de notificación, renovación de medidas cautelares y demás plazos legales— o recibir recordatorios de ellos. Escenarios típicos de activación incluyen, entre otros: mencionar palabras clave como \"plazo de apelación\", \"plazo de presentación de pruebas\", \"plazo de notificación\" o \"plazo de la medida cautelar\"; añadir, modificar o consultar plazos del caso." }, "fr": { "name": "trial-scheduling-and-deadline-monitoring", "summary": "Utilisez cette compétence lorsque l'utilisateur doit suivre les divers délais légaux d'un dossier — calendrier des audiences, calendrier de l'exécution, délais de production des preuves, délais d'appel, délais de signification, renouvellement des mesures conservatoires et autres délais légaux — ou se faire rappeler ceux-ci. Scénarios types de déclenchement, sans s'y limiter : mention de mots-clés tels que « délai d'appel », « délai de production des preuves », « délai de signification » ou « délai de la mesure conservatoire » ; ajout, modification ou consultation d'échéances du dossier." }, "ja": { "name": "trial-scheduling-and-deadline-monitoring", "summary": "ユーザーが事件の公判期日の日程、執行の日程、証拠提出期限、控訴期限、送達期限、保全の更新など各種法定期限を追跡したり、通知を受けたりする必要があるときに使用します。典型的な起動場面（これらに限りません）：「控訴期限」「証拠提出期限」「送達期限」「保全期限」などのキーワードへの言及；事件の期限の追加、変更、照会。" } }, "whenToUse": "Relevant when working on: legal, chinese-law, thuyran.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/THUYRan/Legal-Skills-Chinese" }, { "id": "skill:demo-statement-brief", "slug": "skill-demo-statement-brief", "category": "skill", "marketCategory": null, "tags": [], "canonical": { "brief": "Provides a brief note on earnings-report framing." }, "human": { "zh-CN": { "name": "demo-statement-brief", "summary": "给出一段简短的财报口径提示" }, "en": { "name": "demo-statement-brief", "summary": "Provides a brief note on earnings-report framing." }, "es": { "name": "demo-statement-brief", "summary": "Proporciona una breve nota sobre el encuadre del informe de resultados." }, "fr": { "name": "demo-statement-brief", "summary": "Fournit une brève note sur le cadrage du rapport de résultats." }, "ja": { "name": "demo-statement-brief", "summary": "決算レポートの口径に関する短いメモを提供します。" } }, "whenToUse": "给出一段简短的财报口径提示", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:spider-heal-protocol", "slug": "skill-spider-heal-protocol", "category": "skill", "marketCategory": null, "tags": [], "canonical": { "brief": "External ticket protocol (SUBMIT/STATUS/QUEUE), inbox status file, deduplication and single-writer rules." }, "human": { "zh-CN": { "name": "spider-heal-protocol", "summary": "对外工单协议（SUBMIT/STATUS/QUEUE）、inbox 状态文件、去重与单写者规则" }, "en": { "name": "spider-heal-protocol", "summary": "External ticket protocol (SUBMIT/STATUS/QUEUE), inbox status file, deduplication and single-writer rules." }, "es": { "name": "spider-heal-protocol", "summary": "Protocolo de tickets externo (SUBMIT/STATUS/QUEUE), archivo de estado de inbox, reglas de deduplicación y de escritor único." }, "fr": { "name": "spider-heal-protocol", "summary": "Protocole de tickets externe (SUBMIT/STATUS/QUEUE), fichier d'état inbox, règles de déduplication et d'écrivain unique." }, "ja": { "name": "spider-heal-protocol", "summary": "外部向けチケットプロトコル（SUBMIT/STATUS/QUEUE）、inbox ステータスファイル、重複排除と単一ライターのルール。" } }, "whenToUse": "对外工单协议（SUBMIT/STATUS/QUEUE）、inbox 状态文件、去重与单写者规则", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:spider-heal-repair", "slug": "skill-spider-heal-repair", "category": "skill", "marketCategory": null, "tags": [], "canonical": { "brief": "Repair execution: shallow checkout, ticket reading and triage, targeted fixes, verification chain, opening PRs via curl, credential-redaction discipline." }, "human": { "zh-CN": { "name": "spider-heal-repair", "summary": "修复执行：浅检出、读工单分诊、定向修复、验证链、curl 开 PR、凭据脱敏纪律" }, "en": { "name": "spider-heal-repair", "summary": "Repair execution: shallow checkout, ticket reading and triage, targeted fixes, verification chain, opening PRs via curl, credential-redaction discipline." }, "es": { "name": "spider-heal-repair", "summary": "Ejecución de reparación: checkout superficial, lectura y triaje de tickets, correcciones dirigidas, cadena de verificación, apertura de PR mediante curl, disciplina de enmascaramiento de credenciales." }, "fr": { "name": "spider-heal-repair", "summary": "Exécution de réparation : checkout superficiel, lecture et triage des tickets, corrections ciblées, chaîne de vérification, ouverture de PR via curl, discipline de masquage des identifiants." }, "ja": { "name": "spider-heal-repair", "summary": "修復の実行：シャローチェックアウト、チケット読み込みとトリアージ、的を絞った修正、検証チェーン、curl による PR 作成、資格情報のマスキング規律。" } }, "whenToUse": "修复执行：浅检出、读工单分诊、定向修复、验证链、curl 开 PR、凭据脱敏纪律", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:spider-heal-notify", "slug": "skill-spider-heal-notify", "category": "skill", "marketCategory": null, "tags": [], "canonical": { "brief": "bot_notify usage for the four event types (ticket final state / PR opened / master-switch change / handoff to human) plus the master-switch verification procedure." }, "human": { "zh-CN": { "name": "spider-heal-notify", "summary": "四类事件通知（工单终态/PR开出/总闸变更/转人工）的 bot_notify 用法 + 总闸核对流程" }, "en": { "name": "spider-heal-notify", "summary": "bot_notify usage for the four event types (ticket final state / PR opened / master-switch change / handoff to human) plus the master-switch verification procedure." }, "es": { "name": "spider-heal-notify", "summary": "Uso de bot_notify para los cuatro tipos de eventos (estado final del ticket / PR abierta / cambio del interruptor maestro / derivación a humano) más el procedimiento de verificación del interruptor maestro." }, "fr": { "name": "spider-heal-notify", "summary": "Utilisation de bot_notify pour les quatre types d'événements (état final du ticket / PR ouverte / changement de l'interrupteur général / transfert à un humain) plus la procédure de vérification de l'interrupteur général." }, "ja": { "name": "spider-heal-notify", "summary": "4種類のイベント通知（チケット終了状態/PR作成/マスタースイッチ変更/人間への引き継ぎ）の bot_notify の使い方＋マスタースイッチの確認フロー。" } }, "whenToUse": "四类事件通知（工单终态/PR开出/总闸变更/转人工）的 bot_notify 用法 + 总闸核对流程", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:spider-heal-generate", "slug": "skill-spider-heal-generate", "category": "skill", "marketCategory": null, "tags": [], "canonical": { "brief": "Generation execution: kind=generate greenfield creation of a new unit — scaffold from template, implement data fetching, golden-sample cross-period stability discipline, health_verify loop of at most 3 rounds, open PR from the gen/ branch." }, "human": { "zh-CN": { "name": "spider-heal-generate", "summary": "生成执行：kind=generate greenfield 新建单元——模板起接、实现取数、golden 样本跨期稳定纪律、health_verify 循环 ≤3 轮、gen/ 分支开 PR" }, "en": { "name": "spider-heal-generate", "summary": "Generation execution: kind=generate greenfield creation of a new unit — scaffold from template, implement data fetching, golden-sample cross-period stability discipline, health_verify loop of at most 3 rounds, open PR from the gen/ branch." }, "es": { "name": "spider-heal-generate", "summary": "Ejecución de generación: creación greenfield de una nueva unidad con kind=generate — andamiaje desde plantilla, implementación de la obtención de datos, disciplina de estabilidad entre periodos de las muestras golden, bucle health_verify de como máximo 3 rondas, apertura de PR desde la rama gen/." }, "fr": { "name": "spider-heal-generate", "summary": "Exécution de génération : création greenfield d'une nouvelle unité avec kind=generate — démarrage à partir du modèle, implémentation de la récupération des données, discipline de stabilité inter-périodes des échantillons golden, boucle health_verify d'au plus 3 tours, ouverture de PR depuis la branche gen/." }, "ja": { "name": "spider-heal-generate", "summary": "生成の実行：kind=generate による greenfield の新規ユニット作成——テンプレートからの雛形構築、データ取得の実装、golden サンプルの期間横断安定性の規律、health_verify ループは最大3回、gen/ ブランチから PR を作成。" } }, "whenToUse": "生成执行：kind=generate greenfield 新建单元——模板起接、实现取数、golden 样本跨期稳定纪律、health_verify 循环 ≤3 轮、gen/ 分支开 PR", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:eco-skill-creator", "slug": "skill-eco-skill-creator", "category": "skill", "marketCategory": null, "tags": [ "ecosystem", "anthropics" ], "canonical": { "brief": "[Ecosystem] Create new SKILL.md skill packages using a structured process." }, "human": { "zh-CN": { "name": "eco-skill-creator", "summary": "[生态] 用结构化流程创建新的 SKILL.md 技能包" }, "en": { "name": "eco-skill-creator", "summary": "[Ecosystem] Create new SKILL.md skill packages using a structured process." }, "es": { "name": "eco-skill-creator", "summary": "[Ecosistema] Crea nuevos paquetes de habilidades SKILL.md mediante un proceso estructurado." }, "fr": { "name": "eco-skill-creator", "summary": "[Écosystème] Créer de nouveaux paquets de compétences SKILL.md selon un processus structuré." }, "ja": { "name": "eco-skill-creator", "summary": "[エコシステム] 構造化されたプロセスで新しい SKILL.md スキルパッケージを作成します。" } }, "whenToUse": "Relevant when working on: ecosystem, anthropics.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/anthropics/skills" }, { "id": "skill:eco-mcp-builder", "slug": "skill-eco-mcp-builder", "category": "skill", "marketCategory": null, "tags": [ "ecosystem", "anthropics" ], "canonical": { "brief": "[Ecosystem] Engineering guide and scaffolding for building MCP servers." }, "human": { "zh-CN": { "name": "eco-mcp-builder", "summary": "[生态] 构建 MCP server 的工程指南与脚手架" }, "en": { "name": "eco-mcp-builder", "summary": "[Ecosystem] Engineering guide and scaffolding for building MCP servers." }, "es": { "name": "eco-mcp-builder", "summary": "[Ecosistema] Guía de ingeniería y andamiaje para construir servidores MCP." }, "fr": { "name": "eco-mcp-builder", "summary": "[Écosystème] Guide d'ingénierie et échafaudage pour construire des serveurs MCP." }, "ja": { "name": "eco-mcp-builder", "summary": "[エコシステム] MCPサーバー構築のためのエンジニアリングガイドとスキャフォールド。" } }, "whenToUse": "Relevant when working on: ecosystem, anthropics.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/anthropics/skills" }, { "id": "skill:eco-canvas-design", "slug": "skill-eco-canvas-design", "category": "skill", "marketCategory": null, "tags": [ "ecosystem", "anthropics" ], "canonical": { "brief": "[Ecosystem] Canvas/graphics design workflow." }, "human": { "zh-CN": { "name": "eco-canvas-design", "summary": "[生态] 画布/图形设计工作流" }, "en": { "name": "eco-canvas-design", "summary": "[Ecosystem] Canvas/graphics design workflow." }, "es": { "name": "eco-canvas-design", "summary": "[Ecosistema] Flujo de trabajo de diseño de lienzo/gráficos." }, "fr": { "name": "eco-canvas-design", "summary": "[Écosystème] Flux de travail de conception de canevas/graphiques." }, "ja": { "name": "eco-canvas-design", "summary": "[エコシステム] キャンバス/グラフィックデザインのワークフロー。" } }, "whenToUse": "Relevant when working on: ecosystem, anthropics.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://github.com/anthropics/skills" }, { "id": "skill:doc-studio", "slug": "skill-doc-studio", "category": "skill", "marketCategory": null, "tags": [], "canonical": { "brief": "Document workshop — Word/Excel/PPT business document generation and editing. Use when the user asks to generate reports, spreadsheets, presentations, or document deliverables, or to edit/convert uploaded docx/xlsx/pptx files. Produces finished documents in Chinese-style business layout, with automat" }, "human": { "zh-CN": { "name": "doc-studio", "summary": "文档工坊——Word/Excel/PPT 商务文档生成与编辑。当用户要求生成报告、表格、演示文稿、文档交付物，或要求编辑/转换上传的 docx/xlsx/pptx 文件时使用。产出符合中式商务版式的成稿，生成后自动质检。" }, "en": { "name": "doc-studio", "summary": "Document workshop — Word/Excel/PPT business document generation and editing. Use when the user asks to generate reports, spreadsheets, presentations, or document deliverables, or to edit/convert uploaded docx/xlsx/pptx files. Produces finished documents in Chinese-style business layout, with automatic quality checks after generation." }, "es": { "name": "doc-studio", "summary": "Taller de documentos: generación y edición de documentos de negocio en Word/Excel/PPT. Úselo cuando el usuario pida generar informes, tablas, presentaciones o entregables documentales, o editar/convertir archivos docx/xlsx/pptx cargados. Produce documentos terminados con maquetación de negocio china y control de calidad automático tras la generación." }, "fr": { "name": "doc-studio", "summary": "Atelier de documents : génération et édition de documents professionnels Word/Excel/PPT. À utiliser lorsque l'utilisateur demande de générer rapports, tableaux, présentations ou livrables documentaires, ou de modifier/convertir des fichiers docx/xlsx/pptx téléversés. Produit des documents finis à la mise en page professionnelle chinoise, avec contrôle qualité automatique après génération." }, "ja": { "name": "doc-studio", "summary": "ドキュメント工房 — Word/Excel/PPT ビジネス文書の生成と編集。レポート、表、プレゼンテーション、文書成果物の生成や、アップロードされた docx/xlsx/pptx ファイルの編集・変換を求められたときに使用します。中国式ビジネス体裁の完成稿を出力し、生成後に自動品質チェックを行います。" } }, "whenToUse": "文档工坊——Word/Excel/PPT 商务文档生成与编辑。当用户要求生成报告、表格、演示文稿、文档交付物，或要求编辑/转换上传的 docx/xlsx/pptx 文件时使用。产出符合中式商务版式的成稿，生成后自动质检。", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:case-retrieval-report", "slug": "skill-case-retrieval-report", "category": "skill", "marketCategory": null, "tags": [ "识律", "法律", "finddata" ], "canonical": { "brief": "Similar-case retrieval and adjudication-rule report. Use when the user wants to search similar cases, produce a case retrieval report, look up adjudication rules or tendencies for a type of dispute, run like-case-like-decision analysis, or find precedents supporting a claim. Runs alternating multi-k" }, "human": { "zh-CN": { "name": "case-retrieval-report", "summary": "类案检索与裁判规则报告。当用户要检索类案、出案例检索报告、查某类纠纷的裁判规则或裁判倾向、做同案同判分析、找支持某主张的判例时使用。走裁判文书库 + 指导案例库多组关键词换搜，输出含真实案号、法院、裁判日期与 provenance 的检索报告。" }, "en": { "name": "case-retrieval-report", "summary": "Similar-case retrieval and adjudication-rule report. Use when the user wants to search similar cases, produce a case retrieval report, look up adjudication rules or tendencies for a type of dispute, run like-case-like-decision analysis, or find precedents supporting a claim. Runs alternating multi-keyword searches across the judgment-document library and the guiding-case library, and outputs a retrieval report with real case numbers, courts, adjudication dates, and provenance." }, "es": { "name": "case-retrieval-report", "summary": "Búsqueda de casos similares e informe de reglas de adjudicación. Úselo cuando el usuario quiera buscar casos similares, generar un informe de búsqueda de casos, consultar las reglas o tendencias de adjudicación de un tipo de disputa, hacer análisis de mismo caso/misma decisión o encontrar precedentes que apoyen una afirmación. Ejecuta búsquedas alternadas con varios juegos de palabras clave sobre la biblioteca de sentencias y la de casos guía, y produce un informe con números de expediente reales, tribunales, fechas de adjudicación y provenance." }, "fr": { "name": "case-retrieval-report", "summary": "Recherche de cas similaires et rapport sur les règles de jugement. À utiliser lorsque l'utilisateur souhaite rechercher des cas similaires, produire un rapport de recherche de cas, consulter les règles ou tendances de jugement d'un type de litige, réaliser une analyse même cas/même décision ou trouver des précédents à l'appui d'une thèse. Lance des recherches alternées avec plusieurs jeux de mots-clés sur la base de décisions de justice et celle des cas directeurs, et produit un rapport avec numéros de dossier réels, tribunaux, dates de jugement et provenance." }, "ja": { "name": "case-retrieval-report", "summary": "類似事件の検索と裁判ルールのレポート。類似事件の検索、事例検索レポートの作成、ある種の紛争の裁判ルールや裁判傾向の調査、同案同判分析、主張を支える判例の探索を求められたときに使用します。裁判文書ライブラリと指導案例ライブラリに対して複数組のキーワードで検索を繰り返し、実際の事件番号、裁判所、裁判日付、provenance を含む検索レポートを出力します。" } }, "whenToUse": "Relevant when working on: 识律, 法律, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:civil-nine-step-workflow", "slug": "skill-civil-nine-step-workflow", "category": "skill", "marketCategory": null, "tags": [ "识律", "法律", "finddata" ], "canonical": { "brief": "Conversational version of the nine-step method for civil cases (basis of claim + nine-step element-based adjudication + norm theory). Use when the user describes a case or provides a complaint/defense statement/evidence/judgment and asks to fix the claims, find the basis of claim, analyze defenses, " }, "human": { "zh-CN": { "name": "civil-nine-step-workflow", "summary": "民事案件九步法会话版（请求权基础 + 要件审判九步法 + 规范说）。当用户描述案情或提供起诉状/答辩状/证据/判决书，要求固定诉讼请求、找请求权基础、抗辩分析、要件拆解、类案检索、争点矩阵、举证责任分配、事实认定、裁判预测或法条/案号真实性核验时使用。输出为纯文本会话推演，不写文件系统、不调脚本。" }, "en": { "name": "civil-nine-step-workflow", "summary": "Conversational version of the nine-step method for civil cases (basis of claim + nine-step element-based adjudication + norm theory). Use when the user describes a case or provides a complaint/defense statement/evidence/judgment and asks to fix the claims, find the basis of claim, analyze defenses, break down elements, search similar cases, build an issue matrix, allocate burdens of proof, determine facts, predict the adjudication, or verify the authenticity of statutes/case numbers. Output is plain-text in-session reasoning; it does not write to the file system or call scripts." }, "es": { "name": "civil-nine-step-workflow", "summary": "Versión conversacional del método de nueve pasos para casos civiles (base de la pretensión + método de nueve pasos de juzgamiento por elementos + teoría de la norma). Úselo cuando el usuario describa un caso o aporte demanda/contestación/pruebas/sentencia y pida fijar las pretensiones, hallar la base de la pretensión, analizar defensas, desglosar elementos, buscar casos similares, construir una matriz de cuestiones, repartir la carga de la prueba, determinar hechos, predecir la decisión o verificar la autenticidad de normas/números de expediente. La salida es una deducción en texto plano dentro de la sesión; no escribe en el sistema de archivos ni llama a scripts." }, "fr": { "name": "civil-nine-step-workflow", "summary": "Version conversationnelle de la méthode en neuf étapes pour les affaires civiles (fondement de la demande + méthode en neuf étapes du jugement par éléments + théorie de la norme). À utiliser lorsque l'utilisateur décrit une affaire ou fournit assignation/conclusions/preuves/jugement et demande de fixer les demandes, trouver le fondement de la demande, analyser les moyens de défense, décomposer les éléments, rechercher des cas similaires, construire une matrice des points en litige, répartir la charge de la preuve, qualifier les faits, prédire la décision ou vérifier l'authenticité des textes/numéros de dossier. La sortie est un raisonnement en texte brut dans la session ; aucune écriture sur le système de fichiers, aucun appel de script." }, "ja": { "name": "civil-nine-step-workflow", "summary": "民事事件のための九步法の会話版（請求権基礎＋要件審判九步法＋規範説）。事件の説明や訴状/答弁書/証拠/判決書の提供をもとに、訴訟請求の確定、請求権基礎の特定、抗弁分析、要件の分解、類似事件の検索、争点マトリクス、立証責任の配分、事実認定、裁判予測、法条/事件番号の真偽確認を求められたときに使用します。出力はプレーンテキストによるセッション内推演で、ファイルシステムへの書き込みやスクリプトの呼び出しは行いません。" } }, "whenToUse": "Relevant when working on: 识律, 法律, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:contract-review-cn", "slug": "skill-contract-review-cn", "category": "skill", "marketCategory": null, "tags": [ "识律", "法律", "finddata" ], "canonical": { "brief": "Contract review under Chinese law. After determining that the contract is governed by mainland China law, loads general principles plus five industry review cards (labor/loans/confidentiality/sales/SaaS), reviews clause by clause, and produces a dual-axis risk issue list and a review memo; precise s" }, "human": { "zh-CN": { "name": "contract-review-cn", "summary": "中国法合同审查。判定合同适用中国大陆法后，加载通用原则与五张行业审查卡（劳动/借贷/保密/买卖/SaaS），逐条审查并产出双轴风险问题清单与审查备忘录；涉精确法条、阈值、生效日期时经法律数据工具核验。当用户要求审查/审阅/markup 合同、检查条款、问某条款在中国法下是否有效或如何修改时使用。仅覆盖中国大陆法，不含港澳台、不含刑事与行政、不含外国法规则包。" }, "en": { "name": "contract-review-cn", "summary": "Contract review under Chinese law. After determining that the contract is governed by mainland China law, loads general principles plus five industry review cards (labor/loans/confidentiality/sales/SaaS), reviews clause by clause, and produces a dual-axis risk issue list and a review memo; precise statutes, thresholds, and effective dates are verified through legal data tools. Use when the user asks to review/annotate/mark up a contract, check clauses, or asks whether a clause is valid under Chinese law or how to amend it. Covers mainland China law only — no Hong Kong/Macau/Taiwan, no criminal or administrative law, no foreign-law rule packs." }, "es": { "name": "contract-review-cn", "summary": "Revisión de contratos según el derecho chino. Tras determinar que el contrato se rige por el derecho de la China continental, carga los principios generales y cinco fichas de revisión sectoriales (laboral/préstamos/confidencialidad/compraventa/SaaS), revisa cláusula por cláusula y produce una lista de riesgos de doble eje y una memoria de revisión; las normas exactas, los umbrales y las fechas de entrada en vigor se verifican con herramientas de datos jurídicos. Úselo cuando el usuario pida revisar/anotar/marcar un contrato, examinar cláusulas, o pregunte si una cláusula es válida bajo derecho chino y cómo modificarla. Cubre solo el derecho de la China continental: sin Hong Kong/Macao/Taiwán, sin derecho penal ni administrativo, sin paquetes de reglas de derecho extranjero." }, "fr": { "name": "contract-review-cn", "summary": "Révision de contrats en droit chinois. Après avoir déterminé que le contrat est régi par le droit de la Chine continentale, charge les principes généraux et cinq fiches de révision sectorielles (travail/prêts/confidentialité/vente/SaaS), examine clause par clause et produit une liste de risques à double axe et un mémoire de révision ; les textes exacts, seuils et dates d'entrée en vigueur sont vérifiés via des outils de données juridiques. À utiliser lorsque l'utilisateur demande de réviser/annoter/marquer un contrat, vérifier des clauses, ou demande si une clause est valide en droit chinois et comment la modifier. Couvre uniquement le droit de la Chine continentale : ni Hong Kong/Macao/Taïwan, ni droit pénal ou administratif, ni packs de règles de droit étranger." }, "ja": { "name": "contract-review-cn", "summary": "中国法による契約審査。契約に中国大陸法が適用されると判定した後、一般原則と5枚の業界別審査カード（労働/貸借/秘密保持/売買/SaaS）を読み込み、条項ごとに審査して二軸のリスク課題リストと審査メモを作成します。正確な法条、閾値、発効日については法律データツールで検証します。契約のレビュー/マークアップ、条項のチェック、ある条項が中国法の下で有効か・どう修正すべきかを尋ねられたときに使用します。対象は中国大陸法のみで、港澳台、刑事・行政法、外国法ルールパックは含みません。" } }, "whenToUse": "Relevant when working on: 识律, 法律, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:criminal-defense-workflow", "slug": "skill-criminal-defense-workflow", "category": "skill", "marketCategory": null, "tags": [ "识律", "法律", "finddata" ], "canonical": { "brief": "End-to-end controller for the criminal defense process (entry skill of the 识律 criminal defense suite). Use when the user provides criminal case materials (police prosecution recommendation/indictment/case file/judgment) or asks for case-file analysis, mapping of defense arguments, choice of defense " }, "human": { "zh-CN": { "name": "criminal-defense-workflow", "summary": "刑事辩护全流程总控（识律刑事辩护套件入口技能）。当用户提出刑事案件材料（起诉意见书/起诉书/卷宗/判决书）、要求刑事阅卷分析、辩点梳理、辩护策略路径选择、定性或量刑方向判断时使用。按审查起诉—一审的阶段推进（侦查阶段不在本技能主流程内，见阶段 1 说明），调度套件内「认罪认罚/不起诉论证」与「庭审发问提纲」两个原子技能，并统一承担法条锚定、类案检索与红线校验。" }, "en": { "name": "criminal-defense-workflow", "summary": "End-to-end controller for the criminal defense process (entry skill of the 识律 criminal defense suite). Use when the user provides criminal case materials (police prosecution recommendation/indictment/case file/judgment) or asks for case-file analysis, mapping of defense arguments, choice of defense strategy paths, or judgment on offense characterization or sentencing direction. Progresses through stages from review-for-prosecution to first instance (the investigation stage is outside this skill's main flow — see the stage 1 notes), dispatches the suite's two atomic skills, 'plea leniency / non-prosecution argumentation' and 'court examination outline', and centrally handles statute anchoring, similar-case retrieval, and red-line checks." }, "es": { "name": "criminal-defense-workflow", "summary": "Controlador integral del proceso de defensa penal (skill de entrada de la suite de defensa penal de 识律). Úselo cuando el usuario aporte materiales de un caso penal (dictamen policial de propuesta fiscal/escrito de acusación/expediente/sentencia) o pida análisis del expediente, ordenación de argumentos de defensa, elección de rutas de estrategia defensiva o valoración de la calificación del delito y la orientación de la pena. Avanza por etapas desde la revisión fiscal hasta la primera instancia (la etapa de investigación queda fuera del flujo principal de esta skill; véase la nota de la etapa 1), despacha las dos skills atómicas de la suite, «argumentación de admisión de culpabilidad y no enjuiciamiento» y «guion de interrogatorio en juicio», y asume de forma centralizada el anclaje de normas, la búsqueda de casos similares y la verificación de líneas rojas." }, "fr": { "name": "criminal-defense-workflow", "summary": "Contrôleur de bout en bout du processus de défense pénale (skill d'entrée de la suite de défense pénale 识律). À utiliser lorsque l'utilisateur fournit des documents d'une affaire pénale (avis de poursuite/acte d'accusation/dossier/jugement) ou demande une analyse du dossier, l'identification des moyens de défense, le choix d'un chemin de stratégie défensive, ou une appréciation de la qualification des faits et de l'orientation de la peine. Progresse par étapes de la phase de réquisition à la première instance (la phase d'enquête est hors du flux principal de ce skill ; voir la note de l'étape 1), déclenche les deux skills atomiques de la suite, « argumentation reconnaissance de culpabilité et non-poursuite » et « plan d'interrogatoire à l'audience », et prend en charge de façon centralisée l'ancrage des textes, la recherche de cas similaires et les contrôles de lignes rouges." }, "ja": { "name": "criminal-defense-workflow", "summary": "刑事弁護プロセス全体の統括（识律刑事弁護スイートの入口スキル）。刑事事件の資料（起訴意見書/起訴状/巻宗/判決書）が提示されたとき、閲巻分析、弁護論点の整理、弁護戦略ルートの選択、罪名の性質判断や量刑の方針判断を求められたときに使用します。審査起訴から第一審までの段階で進行し（捜査段階は本スキルの主フロー外。ステージ1の説明を参照）、スイート内の2つのアトミックスキル「認罪認罰・不起訴論証」「法廷尋問提綱」を呼び出し、法条のアンカー付け、類似事件の検索、レッドライン検証を一括して担います。" } }, "whenToUse": "Relevant when working on: 识律, 法律, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:criminal-plea-and-nonprosecution", "slug": "skill-criminal-plea-and-nonprosecution", "category": "skill", "marketCategory": null, "tags": [ "识律", "法律", "finddata" ], "canonical": { "brief": "Plea leniency and non-prosecution argumentation (atomic skill of the 识律 criminal defense suite). For the review-for-prosecution stage: use when the user asks 'should we accept plea leniency', 'plea leniency negotiation', 'sentencing recommendation negotiation', 'can we get non-prosecution', 'statuto" }, "human": { "zh-CN": { "name": "criminal-plea-and-nonprosecution", "summary": "认罪认罚与不起诉论证（识律刑事辩护套件原子技能）。审查起诉阶段使用：当用户问\"要不要认罪认罚\"\"认罪认罚协商\"\"量刑建议协商\"\"能不能不起诉\"\"法定/存疑/酌定不起诉\"或由 criminal-defense-workflow 调度时使用。输出证据态势、利弊权衡、量刑建议预测与协商三层条款、三种不起诉路径论证，并做当事人知情告知。" }, "en": { "name": "criminal-plea-and-nonprosecution", "summary": "Plea leniency and non-prosecution argumentation (atomic skill of the 识律 criminal defense suite). For the review-for-prosecution stage: use when the user asks 'should we accept plea leniency', 'plea leniency negotiation', 'sentencing recommendation negotiation', 'can we get non-prosecution', 'statutory/insufficient-evidence/discretionary non-prosecution', or is dispatched by criminal-defense-workflow. Outputs the evidence posture, a pros-and-cons weighing, a sentencing recommendation forecast with three-tier negotiation clauses, argumentation for the three non-prosecution paths, and informed notifications to the parties." }, "es": { "name": "criminal-plea-and-nonprosecution", "summary": "Argumentación de admisión de culpabilidad y no enjuiciamiento (skill atómica de la suite de defensa penal de 识律). Para la etapa de revisión fiscal: úselo cuando el usuario pregunte «¿conviene aceptar la admisión de culpabilidad?», «negociación de admisión de culpabilidad», «negociación de la propuesta de pena», «¿se puede obtener el no enjuiciamiento?», «no enjuiciamiento legal/por duda/discrecional», o sea despachado por criminal-defense-workflow. Produce el panorama probatorio, la ponderación de pros y contras, la predicción de la propuesta de pena con cláusulas de negociación en tres niveles, la argumentación de las tres vías de no enjuiciamiento y las notificaciones informadas a las partes." }, "fr": { "name": "criminal-plea-and-nonprosecution", "summary": "Argumentation reconnaissance de culpabilité et non-poursuite (skill atomique de la suite de défense pénale 识律). Pour la phase de réquisition : à utiliser lorsque l'utilisateur demande « faut-il accepter la reconnaissance de culpabilité », « négociation de reconnaissance de culpabilité », « négociation de la proposition de peine », « peut-on obtenir une non-poursuite », « non-poursuite légale/pour doute/discrétionnaire », ou est déclenché par criminal-defense-workflow. Produit la posture probatoire, la pesée des avantages et inconvénients, la prévision de la proposition de peine avec des clauses de négociation à trois niveaux, l'argumentation des trois voies de non-poursuite et l'information éclairée des parties." }, "ja": { "name": "criminal-plea-and-nonprosecution", "summary": "認罪認罰と不起訴の論証（识律刑事弁護スイートのアトミックスキル）。審査起訴段階で使用：「認罪認罰を受け入れるべきか」「認罪認罰の協議」「量刑意見の協議」「不起訴にできるか」「法定/存疑/酌定不起訴」などと尋ねられたとき、または criminal-defense-workflow から呼び出されたときに使用します。証拠の状況、メリット・デメリットの比較衡量、量刑意見の予測と三層の協議条項、3つの不起訴ルートの論証を出力し、当事者への知情告知も行います。" } }, "whenToUse": "Relevant when working on: 识律, 法律, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:criminal-trial-examination-outline", "slug": "skill-criminal-trial-examination-outline", "category": "skill", "marketCategory": null, "tags": [ "识律", "法律", "finddata" ], "canonical": { "brief": "Court examination and evidence-challenge outlines (atomic skill of the 识律 criminal defense suite). For the first-instance stage: use when the user says 'pre-trial conference', 'suppression application', 'exclusion of illegal evidence', 'court examination', 'examination outline', 'evidence-challenge " }, "human": { "zh-CN": { "name": "criminal-trial-examination-outline", "summary": "庭审发问与举证质证提纲（识律刑事辩护套件原子技能）。一审阶段使用：当用户说\"庭前会议\"\"排非申请\"\"非法证据排除\"\"庭审发问\"\"发问提纲\"\"质证提纲\"\"举证质证\"\"交叉询问\"\"逐份质证\"\"辩方举证\"或由 criminal-defense-workflow 调度时使用。输出庭前会议议程应对、非法证据排除申请、公诉人发问预判与应对、辩护人对被告人/证人/鉴定人的发问提纲、辩方举证提纲与逐份质证意见（书面版 + 庭上口述版）。" }, "en": { "name": "criminal-trial-examination-outline", "summary": "Court examination and evidence-challenge outlines (atomic skill of the 识律 criminal defense suite). For the first-instance stage: use when the user says 'pre-trial conference', 'suppression application', 'exclusion of illegal evidence', 'court examination', 'examination outline', 'evidence-challenge outline', 'evidence presentation and challenge', 'cross-examination', 'document-by-document challenge', 'defense presentation of evidence', or is dispatched by criminal-defense-workflow. Outputs pre-trial conference agenda responses, illegal-evidence exclusion applications, anticipation of and responses to prosecutor questions, defense examination outlines for the defendant/witnesses/expert witnesses, and the defense evidence plan with document-by-document challenge opinions (written version + courtroom spoken version)." }, "es": { "name": "criminal-trial-examination-outline", "summary": "Guiones de interrogatorio en juicio y de impugnación de pruebas (skill atómica de la suite de defensa penal de 识律). Para la primera instancia: úselo cuando el usuario diga «conferencia previa al juicio», «solicitud de exclusión», «exclusión de prueba ilícita», «interrogatorio en juicio», «guion de interrogatorio», «guion de impugnación», «presentación e impugnación de pruebas», «contrainterrogatorio», «impugnación documento por documento», «prueba de la defensa», o sea despachado por criminal-defense-workflow. Produce respuestas para el orden del día de la conferencia previa, solicitudes de exclusión de prueba ilícita, previsiones y respuestas ante las preguntas del fiscal, guiones de interrogatorio de la defensa para el acusado/los testigos/los peritos, y el plan de prueba de la defensa con impugnaciones documento por documento (versión escrita + versión oral para el juicio)." }, "fr": { "name": "criminal-trial-examination-outline", "summary": "Plans d'interrogatoire à l'audience et de discussion des preuves (skill atomique de la suite de défense pénale 识律). Pour la première instance : à utiliser lorsque l'utilisateur dit « conférence préparatoire », « demande d'exclusion », « exclusion des preuves illicites », « interrogatoire à l'audience », « plan d'interrogatoire », « plan de discussion des preuves », « production et discussion des preuves », « contre-interrogatoire », « discussion pièce par pièce », « preuves de la défense », ou est déclenché par criminal-defense-workflow. Produit les réponses à l'ordre du jour de la conférence préparatoire, les demandes d'exclusion de preuves illicites, l'anticipation des questions du ministère public et les réponses à y apporter, les plans d'interrogatoire de la défense pour l'accusé/les témoins/les experts, ainsi que le plan de preuve de la défense avec des observations pièce par pièce (version écrite + version orale pour l'audience)." }, "ja": { "name": "criminal-trial-examination-outline", "summary": "法廷尋問と挙証・質証の提綱（识律刑事弁護スイートのアトミックスキル）。第一審段階で使用：「庭前会議」「排非申請」「違法収集証拠の排除」「法廷尋問」「尋問提綱」「質証提綱」「挙証質証」「交互尋問」「一件ずつの質証」「弁護側挙証」と言われたとき、または criminal-defense-workflow から呼び出されたときに使用します。庭前会議の議題対応、違法証拠排除申立て、検察官の尋問の予測と対応、弁護人による被告人/証人/鑑定人への尋問提綱、弁護側挙証提綱と一件ずつの質証意見（書面版＋法廷口頭版）を出力します。" } }, "whenToUse": "Relevant when working on: 识律, 法律, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:evidence-card-organizer", "slug": "skill-evidence-card-organizer", "category": "skill", "marketCategory": null, "tags": [ "识律", "法律", "finddata" ], "canonical": { "brief": "Evidence card extraction and material list organization (case-filing material organization skill). Use when the user submits a batch of case materials (contracts, transfer records, chat logs, receipts, proofs of service, etc.) and asks to extract evidence, compile an evidence list, organize material" }, "human": { "zh-CN": { "name": "evidence-card-organizer", "summary": "证据卡片提取与材料清单整理（立案材料整理技能）。当用户提交一批案件材料（合同、转账记录、聊天记录、票据、送达回证等），要求提取证据、编证据目录、整理材料清单或做证据关系与矛盾排查时使用。输出双视图证据卡片库与待补强清单，供起诉状/答辩状起草直接引用。" }, "en": { "name": "evidence-card-organizer", "summary": "Evidence card extraction and material list organization (case-filing material organization skill). Use when the user submits a batch of case materials (contracts, transfer records, chat logs, receipts, proofs of service, etc.) and asks to extract evidence, compile an evidence list, organize material inventories, or screen evidence relationships and contradictions. Outputs a dual-view evidence card library and a to-strengthen list, ready for direct citation when drafting complaints/defense statements." }, "es": { "name": "evidence-card-organizer", "summary": "Extracción de fichas de prueba y organización de la lista de materiales (skill de organización de materiales para la presentación del caso). Úselo cuando el usuario entregue un lote de materiales del caso (contratos, registros de transferencias, historiales de chat, recibos, constancias de notificación, etc.) y pida extraer pruebas, elaborar el índice de pruebas, organizar listas de materiales o detectar relaciones y contradicciones entre las pruebas. Produce una biblioteca de fichas de prueba de doble vista y una lista de elementos por reforzar, listos para citarse directamente al redactar la demanda/la contestación." }, "fr": { "name": "evidence-card-organizer", "summary": "Extraction de fiches de preuves et organisation de la liste des pièces (skill d'organisation des pièces pour la saisine). À utiliser lorsque l'utilisateur soumet un lot de pièces (contrats, relevés de virements, historiques de messagerie, reçus, avis de signification, etc.) et demande d'extraire les preuves, dresser l'inventaire des preuves, organiser les listes de pièces ou repérer les liens et contradictions entre preuves. Produit une bibliothèque de fiches de preuves à double vue et une liste des éléments à renforcer, directement citables lors de la rédaction de l'assignation/des conclusions." }, "ja": { "name": "evidence-card-organizer", "summary": "証拠カードの抽出と資料リストの整理（立案件資料の整理スキル）。契約書、送金記録、チャット記録、領収書、送達回証などの事件資料一式が提出され、証拠の抽出、証拠目録の作成、資料リストの整理、証拠間の関係と矛盾の点検を求められたときに使用します。2つのビューを持つ証拠カードライブラリと補強必要リストを出力し、訴状/答弁書の起草時に直接引用できます。" } }, "whenToUse": "Relevant when working on: 识律, 法律, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:labor-arbitration-workflow", "slug": "skill-labor-arbitration-workflow", "category": "skill", "marketCategory": null, "tags": [ "识律", "法律", "finddata" ], "canonical": { "brief": "Full labor dispute arbitration process (entry skill of the 识律 labor law suite). Use when the user describes a labor dispute (unlawful termination/economic compensation/overtime pay/double wages for no written contract/transfer with pay cut/work-injury benefits), asks for an arbitration application o" }, "human": { "zh-CN": { "name": "labor-arbitration-workflow", "summary": "劳动争议仲裁全流程（识律劳动法套件入口技能）。当用户描述劳动争议案情（违法解除/经济补偿/加班费/未签合同二倍工资/调岗降薪/工伤待遇）、要求仲裁申请书或答辩书、证据目录、一裁终局判断、仲裁时限表，或以裁判者身份要求争议焦点归纳与裁决说理时使用。按立场（申请人/被申请人/裁判者）推进八阶段流程，并按需点名套件内两个计算器技能。" }, "en": { "name": "labor-arbitration-workflow", "summary": "Full labor dispute arbitration process (entry skill of the 识律 labor law suite). Use when the user describes a labor dispute (unlawful termination/economic compensation/overtime pay/double wages for no written contract/transfer with pay cut/work-injury benefits), asks for an arbitration application or defense statement, an evidence list, a finality determination (one award, final), an arbitration deadline table, or — as the adjudicator — asks for issues to be summarized and award reasoning drafted. Advances an eight-stage process by position (claimant/respondent/adjudicator) and calls the suite's two calculator skills as needed." }, "es": { "name": "labor-arbitration-workflow", "summary": "Proceso completo de arbitraje de controversias laborales (skill de entrada de la suite de derecho laboral de 识律). Úselo cuando el usuario describa una disputa laboral (despido ilegal/indemnización por cese/horas extra/doble salario por falta de contrato escrito/traslado con rebaja salarial/prestaciones por accidente de trabajo), pida una solicitud de arbitraje o una contestación, un índice de pruebas, una determinación de firmeza (laudo en única instancia), un calendario de plazos arbitrales, o —en calidad de árbitro— pida sintetizar los puntos en disputa y redactar la motivación del laudo. Desarrolla un proceso de ocho etapas según la posición (solicitante/demandado/árbitro) y llama a las dos skills de cálculo de la suite según necesidad." }, "fr": { "name": "labor-arbitration-workflow", "summary": "Processus complet d'arbitrage des litiges du travail (skill d'entrée de la suite de droit du travail de 识律). À utiliser lorsque l'utilisateur décrit un litige du travail (licenciement illégal/indemnité de rupture/heures supplémentaires/double salaire faute d'écrit/mutation avec baisse de salaire/prestations d'accident du travail), demande une requête d'arbitrage ou une défense, un inventaire des preuves, une appréciation du caractère définitif (une seule instance), un tableau des délais d'arbitrage, ou — en qualité d'arbitre — la synthèse des points en litige et la motivation de la sentence. Déroule un processus en huit étapes selon la position (demandeur/défendeur/arbitre) et appelle au besoin les deux skills de calcul de la suite." }, "ja": { "name": "labor-arbitration-workflow", "summary": "労働争議仲裁の全プロセス（识律労働法スイートの入口スキル）。労働争議の案情（違法解除/経済補償/残業代/書面契約未締結の二倍賃金/配置転換・減給/労災給付）の説明、仲裁申立書や答弁書、証拠目録、一裁終局の判断、仲裁期限表を求められたとき、または裁判者として争点の整理と裁決の説理を求められたときに使用します。立場（申立人/被申立人/裁判者）に応じて八段階のプロセスを進め、必要に応じてスイート内の2つの計算機スキルを呼び出します。" } }, "whenToUse": "Relevant when working on: 识律, 法律, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:labor-compensation-calculator", "slug": "skill-labor-compensation-calculator", "category": "skill", "marketCategory": null, "tags": [ "识律", "法律", "finddata" ], "canonical": { "brief": "Labor compensation and severance calculator (atomic skill of the 识律 labor law suite, one of the two calculators). Use when the user asks to compute severance N for contract termination, unlawful-termination damages 2N, pay in lieu of notice N+1, double-wage differentials for a missing written labor " }, "human": { "zh-CN": { "name": "labor-compensation-calculator", "summary": "劳动补偿金与经济补偿金计算器（识律劳动法套件原子技能，双计算器之一）。当用户要求计算解除劳动合同经济补偿 N、违法解除赔偿金 2N、代通知金 N+1、未签书面劳动合同二倍工资差额、加班费、未休年休假工资报酬，或需要判断三倍封顶（社平工资三倍 × 12 年上限）时使用。法律依据走识律条级解析，数值基数走柏讯统计源，每个结果强制附公式与基数来源。" }, "en": { "name": "labor-compensation-calculator", "summary": "Labor compensation and severance calculator (atomic skill of the 识律 labor law suite, one of the two calculators). Use when the user asks to compute severance N for contract termination, unlawful-termination damages 2N, pay in lieu of notice N+1, double-wage differentials for a missing written labor contract, overtime pay, or pay for unused annual leave, or needs to apply the triple cap (3x the average social wage × 12-year ceiling). Legal bases come from 识律 article-level parsing; numeric bases come from the 柏讯 statistics source; every result must carry its formula and base source." }, "es": { "name": "labor-compensation-calculator", "summary": "Calculadora de compensaciones laborales y de cese (skill atómica de la suite de derecho laboral de 识律, una de las dos calculadoras). Úselo cuando el usuario pida calcular la indemnización N por extinción del contrato, los daños 2N por despido ilegal, el sustitutivo del preaviso N+1, los diferenciales de doble salario por falta de contrato escrito, las horas extra o la retribución de vacaciones anuales no disfrutadas, o necesite aplicar el tope del triple (3 veces el salario social medio × tope de 12 años). Las bases legales provienen del análisis artículo a artículo de 识律; las bases numéricas, de la fuente estadística de 柏讯; cada resultado lleva obligatoriamente su fórmula y la fuente de la base." }, "fr": { "name": "labor-compensation-calculator", "summary": "Calculateur d'indemnités de rupture (skill atomique de la suite de droit du travail de 识律, l'un des deux calculateurs). À utiliser lorsque l'utilisateur demande de calculer l'indemnité N de rupture du contrat, les dommages-intérêts 2N pour licenciement illégal, l'indemnité de préavis N+1, les différentiels de double salaire en l'absence de contrat écrit, les heures supplémentaires ou la rémunération des congés annuels non pris, ou doit appliquer le plafond du triple (3 fois le salaire social moyen × plafond de 12 ans). Les bases juridiques passent par l'analyse article par article de 识律 ; les bases chiffrées par la source statistique de 柏讯 ; chaque résultat comporte obligatoirement sa formule et la source de la base." }, "ja": { "name": "labor-compensation-calculator", "summary": "労働補償金・経済補償金の計算機（识律労働法スイートのアトミックスキル、2つの計算機のうちの1つ）。契約解除の経済補償N、違法解除の賠償金2N、代通知金N+1、書面労働契約未締結の二倍賃金差額、残業代、未消化年次休暇の賃金報酬の計算、または三倍封頂（社会平均賃金の3倍 × 12年の上限）の判断を求められたときに使用します。法的根拠は识律の条項単位解析、数値基数は柏讯の統計ソースから取得し、各結果には公式と基数の出典を必ず添付します。" } }, "whenToUse": "Relevant when working on: 识律, 法律, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:legal-concept-map", "slug": "skill-legal-concept-map", "category": "skill", "marketCategory": null, "tags": [ "识律", "法律", "finddata" ], "canonical": { "brief": "Structured diagramming of legal concepts. Use when the user wants an explanation of a legal concept or institution ('what is good-faith acquisition', 'how is apparent agency determined', 'conditions for applying change of circumstances'), wants constituent elements and exceptions broken down, or wan" }, "human": { "zh-CN": { "name": "legal-concept-map", "summary": "法律概念结构化图解。当用户要解释某个法律概念/制度（\"什么是善意取得\"\"表见代理怎么认定\"\"情势变更的适用条件\"）、要拆构成要件与例外、要找关联条文与权威解读时使用。以条级引用解析锚定条文，以五源语料补权威解读，输出要件—例外—关联条文的结构化图解。" }, "en": { "name": "legal-concept-map", "summary": "Structured diagramming of legal concepts. Use when the user wants an explanation of a legal concept or institution ('what is good-faith acquisition', 'how is apparent agency determined', 'conditions for applying change of circumstances'), wants constituent elements and exceptions broken down, or wants related provisions and authoritative interpretations. Anchors provisions via article-level citation parsing, supplements authoritative interpretations from the five-source corpus, and outputs a structured map of elements—exceptions—related provisions." }, "es": { "name": "legal-concept-map", "summary": "Diagramación estructurada de conceptos jurídicos. Úselo cuando el usuario quiera que se le explique un concepto o institución jurídica («¿qué es la adquisición de buena fe?», «¿cómo se determina la apariencia de mandato?», «¿cuándo se aplica la alteración de circunstancias?»), desglosar elementos constitutivos y excepciones, o encontrar disposiciones relacionadas e interpretaciones autorizadas. Ancla las disposiciones mediante análisis de citas a nivel de artículo, complementa las interpretaciones autorizadas con el corpus de cinco fuentes y produce un mapa estructurado de elementos—excepciones—disposiciones relacionadas." }, "fr": { "name": "legal-concept-map", "summary": "Schématisation structurée des concepts juridiques. À utiliser lorsque l'utilisateur veut une explication d'un concept ou d'une institution juridique (« qu'est-ce que l'acquisition de bonne foi », « comment déterminer l'apparence de mandat », « conditions d'application de l'imprévision »), une décomposition des éléments constitutifs et exceptions, ou la recherche de dispositions liées et d'interprétations autorisées. Ancre les dispositions via l'analyse des citations au niveau de l'article, complète les interprétations autorisées avec le corpus à cinq sources, et produit une schématisation structurée éléments—exceptions—dispositions liées." }, "ja": { "name": "legal-concept-map", "summary": "法律概念の構造化ダイアグラム。法律概念・制度の説明（「善意取得とは何か」「表見代理はどう認定するか」「情事変更の適用条件」）、構成要件と例外の分解、関連条文や権威ある解釈の探索を求められたときに使用します。条単位の引用解析で条文をアンカー付けし、五ソースのコーパスで権威ある解釈を補い、要件—例外—関連条文の構造化ダイアグラムを出力します。" } }, "whenToUse": "Relevant when working on: 识律, 法律, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:litigation-drafting", "slug": "skill-litigation-drafting", "category": "skill", "marketCategory": null, "tags": [ "识律", "法律", "finddata" ], "canonical": { "brief": "Civil and commercial litigation document drafting (complaints/defense statements/evidence lists). Use when the user provides case facts or case materials and asks to draft a complaint, a defense statement, key points of an appellate brief, an evidence list, or supporting case-filing documents. Statu" }, "human": { "zh-CN": { "name": "litigation-drafting", "summary": "民商事诉讼文书起草（起诉状/答辩状/证据目录）。当用户提供案情或案件材料，要求写起诉状、答辩状、上诉状要点、证据目录或立案配套文书时使用。法条锚定与类案参照均走真实检索数据源，律师确认要点后才生成成稿。" }, "en": { "name": "litigation-drafting", "summary": "Civil and commercial litigation document drafting (complaints/defense statements/evidence lists). Use when the user provides case facts or case materials and asks to draft a complaint, a defense statement, key points of an appellate brief, an evidence list, or supporting case-filing documents. Statute anchoring and similar-case referencing both go through real retrieval data sources; the finished draft is generated only after the lawyer confirms the key points." }, "es": { "name": "litigation-drafting", "summary": "Redacción de escritos de litigio civil y mercantil (demandas/contestaciones/índices de pruebas). Úselo cuando el usuario aporte los hechos o materiales del caso y pida redactar una demanda, una contestación, los puntos clave de la apelación, un índice de pruebas o documentos complementarios de presentación. El anclaje de normas y la referencia a casos similares usan fuentes de datos de búsqueda reales; el escrito final solo se genera tras la confirmación de los puntos clave por el abogado." }, "fr": { "name": "litigation-drafting", "summary": "Rédaction d'actes de procédure civile et commerciale (assignations/conclusions/inventaires de preuves). À utiliser lorsque l'utilisateur fournit les faits ou les pièces de l'affaire et demande de rédiger une assignation, des conclusions, les points clés de l'appel, un inventaire des preuves ou des pièces annexes de saisine. L'ancrage des textes et le renvoi à des cas similaires passent tous deux par de vraies sources de recherche ; le document final n'est généré qu'après confirmation des points clés par l'avocat." }, "ja": { "name": "litigation-drafting", "summary": "民商事訴訟文書の起草（訴状/答弁書/証拠目録）。案情や事件資料が提供され、訴状、答弁書、控訴状の要点、証拠目録、立案件に付随する文書の作成を求められたときに使用します。法条のアンカー付けと類似事件の参照はいずれも実検索データソースを使用し、弁護士の要点確認後に完成稿を生成します。" } }, "whenToUse": "Relevant when working on: 识律, 法律, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:personal-injury-calculator", "slug": "skill-personal-injury-calculator", "category": "skill", "marketCategory": null, "tags": [ "识律", "法律", "finddata" ], "canonical": { "brief": "Personal injury damages calculator (atomic skill of the 识律 labor law suite, one of the two calculators). Use when the user asks to compute personal injury compensation items — medical expenses, lost earnings, nursing fees, transportation costs, hospital meal allowances, nutrition fees, disability co" }, "human": { "zh-CN": { "name": "personal-injury-calculator", "summary": "人身损害赔偿计算器（识律劳动法套件原子技能，双计算器之一）。当用户要求计算人身损害赔偿项目时使用——医疗费、误工费、护理费、交通费、住院伙食补助费、营养费、残疾赔偿金、残疾辅助器具费、丧葬费、死亡赔偿金、被扶养人生活费、精神损害抚慰金。法律依据走识律条级解析，残疾/死亡赔偿金等基数走柏讯统计源，每个结果强制附公式与基数来源。" }, "en": { "name": "personal-injury-calculator", "summary": "Personal injury damages calculator (atomic skill of the 识律 labor law suite, one of the two calculators). Use when the user asks to compute personal injury compensation items — medical expenses, lost earnings, nursing fees, transportation costs, hospital meal allowances, nutrition fees, disability compensation, disability assistive device costs, funeral expenses, death compensation, dependents' living expenses, and mental distress damages. Legal bases come from 识律 article-level parsing; bases for disability/death compensation etc. come from the 柏讯 statistics source; every result must carry its formula and base source." }, "es": { "name": "personal-injury-calculator", "summary": "Calculadora de indemnización por daños personales (skill atómica de la suite de derecho laboral de 识律, una de las dos calculadoras). Úselo cuando el usuario pida calcular conceptos de indemnización por daños personales: gastos médicos, pérdida de ingresos, cuidados, transporte, manutención en hospital, alimentación especial, indemnización por discapacidad, dispositivos de asistencia, gastos funerarios, indemnización por fallecimiento, gastos de manutención de dependientes y compensación por daño moral. Las bases legales provienen del análisis artículo a artículo de 识律; las bases de la indemnización por discapacidad/fallecimiento etc., de la fuente estadística de 柏讯; cada resultado lleva obligatoriamente su fórmula y la fuente de la base." }, "fr": { "name": "personal-injury-calculator", "summary": "Calculateur d'indemnisation des dommages corporels (skill atomique de la suite de droit du travail de 识律, l'un des deux calculateurs). À utiliser lorsque l'utilisateur demande de calculer les postes d'indemnisation des dommages corporels — frais médicaux, perte de revenus, frais d'assistance, transport, restauration à l'hôpital, frais de nutrition, indemnité d'invalidité, dispositifs d'assistance, frais funéraires, indemnité de décès, frais d'entretien des personnes à charge et préjudice moral. Les bases juridiques passent par l'analyse article par article de 识律 ; les bases de l'indemnité d'invalidité/décès etc. par la source statistique de 柏讯 ; chaque résultat comporte obligatoirement sa formule et la source de la base." }, "ja": { "name": "personal-injury-calculator", "summary": "人身損害賠償の計算機（识律労働法スイートのアトミックスキル、2つの計算機のうちの1つ）。医療費、休業損害、看護費、交通費、入院食事補助費、栄養費、障害賠償金、障害補助具費、葬儀費、死亡賠償金、被扶養者生活費、精神損害撫慰金など人身損害賠償項目の計算を求められたときに使用します。法的根拠は识律の条項単位解析、障害/死亡賠償金などの基数は柏讯の統計ソースから取得し、各結果には公式と基数の出典を必ず添付します。" } }, "whenToUse": "Relevant when working on: 识律, 法律, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:statute-validity-check", "slug": "skill-statute-validity-check", "category": "skill", "marketCategory": null, "tags": [ "识律", "法律", "finddata" ], "canonical": { "brief": "Statute retrieval and validity verification. Use when the user wants the original text of a statute, to check whether a provision is currently in force, to confirm whether article numbers changed in a revision, or to find the legal basis for a legal question ('is it allowed', 'is it valid', 'how is " }, "human": { "zh-CN": { "name": "statute-validity-check", "summary": "法条检索与效力核查。当用户要查某法条原文、核对某条文是否现行有效、确认条号是否因修订变动、查某法律问题的法律依据（\"能不能\"\"是否有效\"\"如何认定\"\"依据哪一条\"）时使用。语义召回候选条文后逐条锚定效力状态，输出带效力状态与 provenance 的法条清单；非\"有效\"版本明确标注不得作现行法。" }, "en": { "name": "statute-validity-check", "summary": "Statute retrieval and validity verification. Use when the user wants the original text of a statute, to check whether a provision is currently in force, to confirm whether article numbers changed in a revision, or to find the legal basis for a legal question ('is it allowed', 'is it valid', 'how is it determined', 'which article applies'). After semantically recalling candidate provisions, each is anchored to its validity status; outputs a statute list with validity status and provenance; versions not 'in force' are explicitly marked as not to be used as current law." }, "es": { "name": "statute-validity-check", "summary": "Búsqueda de normas y verificación de vigencia. Úselo cuando el usuario quiera el texto original de una norma, comprobar si una disposición está vigente, confirmar si la numeración cambió tras una reforma, o encontrar la base jurídica de una cuestión («¿se puede?», «¿es válido?», «¿cómo se determina?», «¿qué artículo aplica?»). Tras recuperar semánticamente las disposiciones candidatas, se ancla el estado de vigencia de cada una; produce una lista de normas con estado de vigencia y provenance; las versiones no «vigentes» se marcan explícitamente como no utilizables como derecho vigente." }, "fr": { "name": "statute-validity-check", "summary": "Recherche de textes et vérification de leur validité. À utiliser lorsque l'utilisateur veut le texte original d'une disposition, vérifier si un article est actuellement en vigueur, confirmer si la numérotation a changé après révision, ou trouver la base juridique d'une question (« est-ce permis », « est-ce valide », « comment est-ce déterminé », « quel article s'applique »). Après rappel sémantique des articles candidats, chacun est ancré à son statut de validité ; produit une liste de dispositions avec statut de validité et provenance ; les versions non « en vigueur » sont explicitement marquées comme inutilisables comme droit actuel." }, "ja": { "name": "statute-validity-check", "summary": "法条の検索と効力検証。ある法条の原文を調べる、条文が現行有効か確認する、改正で条番号が変わったか確認する、法律問題の法的根拠を調べる（「できるか」「有効か」「どう認定するか」「何条に依拠するか」）ときに使用します。候補条文をセマンティックに呼び出した後、1条ずつ効力状態をアンカー付けし、効力状態と provenance 付きの法条リストを出力します。「有効」でない版は現行法として使用しないことを明示します。" } }, "whenToUse": "Relevant when working on: 识律, 法律, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:mission-protocol", "slug": "skill-mission-protocol", "category": "skill", "marketCategory": null, "tags": [], "canonical": { "brief": "Brief parsing; GET polling / POST write-back with five kinds; idempotency-key discipline; a state-machine map; and a per-turn action table." }, "human": { "zh-CN": { "name": "mission-protocol", "summary": "brief 解析、GET 轮询/POST 回写五 kind、幂等键纪律、状态机地图与每回合动作表" }, "en": { "name": "mission-protocol", "summary": "Brief parsing; GET polling / POST write-back with five kinds; idempotency-key discipline; a state-machine map; and a per-turn action table." }, "es": { "name": "mission-protocol", "summary": "Análisis del brief; sondeo GET / escritura POST de cinco tipos; disciplina de claves de idempotencia; mapa de máquina de estados; y tabla de acciones por turno." }, "fr": { "name": "mission-protocol", "summary": "Analyse du brief ; interrogation GET / écriture POST de cinq kind ; discipline des clés d'idempotence ; carte de la machine à états ; et table d'actions par tour." }, "ja": { "name": "mission-protocol", "summary": "brief の解析、GET ポーリング/POST 書き戻し（5種類の kind）、冪等鍵の規律、ステートマシンマップ、ターンごとのアクション表。" } }, "whenToUse": "brief 解析、GET 轮询/POST 回写五 kind、幂等键纪律、状态机地图与每回合动作表", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:mission-planning", "slug": "skill-mission-planning", "category": "skill", "marketCategory": null, "tags": [], "canonical": { "brief": "Goal to a plan of at most 8 steps; step agents must be actually looked up via search_agents, and if none is found execute yourself; the plan, once made, is a commitment — no additions or removals afterwards." }, "human": { "zh-CN": { "name": "mission-planning", "summary": "目标→≤8 步计划；步骤 agent 必须 search_agents 实查，查无自行执行；计划即承诺不再增删" }, "en": { "name": "mission-planning", "summary": "Goal to a plan of at most 8 steps; step agents must be actually looked up via search_agents, and if none is found execute yourself; the plan, once made, is a commitment — no additions or removals afterwards." }, "es": { "name": "mission-planning", "summary": "Objetivo a un plan de 8 pasos como máximo; los agentes de paso deben buscarse realmente con search_agents y, si no hay ninguno, ejecutarlo uno mismo; el plan, una vez hecho, es un compromiso: no se añade ni se elimina nada después." }, "fr": { "name": "mission-planning", "summary": "Objectif vers un plan de 8 étapes au maximum ; les agents d'étape doivent être réellement recherchés via search_agents, et à défaut exécuter soi-même ; le plan, une fois établi, est un engagement — plus d'ajouts ni de suppressions ensuite." }, "ja": { "name": "mission-planning", "summary": "目標から8ステップ以内の計画へ。ステップの agent は必ず search_agents で実査し、見つからなければ自ら実行。計画は一度立てたらコミットメントであり、後からの追加・削除はしない。" } }, "whenToUse": "目标→≤8 步计划；步骤 agent 必须 search_agents 实查，查无自行执行；计划即承诺不再增删", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:mission-execution", "slug": "skill-mission-execution", "category": "skill", "marketCategory": null, "tags": [], "canonical": { "brief": "Serial execution of steps with status write-back; window/budget/retry discipline; fall back to self-execution when delegation is unavailable." }, "human": { "zh-CN": { "name": "mission-execution", "summary": "步骤串行执行与状态回写；窗口/预算/重试纪律；委派不可用转自执行" }, "en": { "name": "mission-execution", "summary": "Serial execution of steps with status write-back; window/budget/retry discipline; fall back to self-execution when delegation is unavailable." }, "es": { "name": "mission-execution", "summary": "Ejecución en serie de los pasos con escritura de estado; disciplina de ventana/presupuesto/reintentos; si la delegación no está disponible, pasar a la autoejecución." }, "fr": { "name": "mission-execution", "summary": "Exécution sérielle des étapes avec réécriture du statut ; discipline fenêtre/budget/retry ; si la délégation est indisponible, passer en auto-exécution." }, "ja": { "name": "mission-execution", "summary": "ステップの逐次実行とステータスの書き戻し。ウィンドウ/予算/リトライの規律。委譲が使えない場合は自己実行に切り替え。" } }, "whenToUse": "步骤串行执行与状态回写；窗口/预算/重试纪律；委派不可用转自执行", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:mission-reporting", "slug": "skill-mission-reporting", "category": "skill", "marketCategory": null, "tags": [], "canonical": { "brief": "Final-state reports (complete/partial delivery) and bot_notify notification discipline." }, "human": { "zh-CN": { "name": "mission-reporting", "summary": "终态报告（完整/部分交付）与 bot_notify 通知纪律" }, "en": { "name": "mission-reporting", "summary": "Final-state reports (complete/partial delivery) and bot_notify notification discipline." }, "es": { "name": "mission-reporting", "summary": "Informes de estado final (entrega completa/parcial) y disciplina de notificación con bot_notify." }, "fr": { "name": "mission-reporting", "summary": "Rapports d'état final (livraison complète/partielle) et discipline de notification via bot_notify." }, "ja": { "name": "mission-reporting", "summary": "最終状態レポート（完全/部分納品）と bot_notify 通知の規律。" } }, "whenToUse": "终态报告（完整/部分交付）与 bot_notify 通知纪律", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:screenplay-structure-analysis", "slug": "skill-screenplay-structure-analysis", "category": "skill", "marketCategory": null, "tags": [ "文本引擎", "剧本", "finddata" ], "canonical": { "brief": "Script structure analysis: maps the act/scene (zhe/chu) skeleton and plot progression of public-domain classic dramas. Data is retrieved in real time from the fd-text-core engine's script corpus (preset: structure)." }, "human": { "zh-CN": { "name": "screenplay-structure-analysis", "summary": "剧本结构分析：公版经典戏剧的幕场/折出骨架与情节推进梳理。数据经 fd-text-core 引擎的剧本语料实时检索（预设 structure）。" }, "en": { "name": "screenplay-structure-analysis", "summary": "Script structure analysis: maps the act/scene (zhe/chu) skeleton and plot progression of public-domain classic dramas. Data is retrieved in real time from the fd-text-core engine's script corpus (preset: structure)." }, "es": { "name": "screenplay-structure-analysis", "summary": "Análisis de la estructura dramática: esquematiza el esqueleto de actos/escenas (zhe/chu) y la progresión argumental de dramas clásicos de dominio público. Los datos se recuperan en tiempo real del corpus de guiones del motor fd-text-core (preset: structure)." }, "fr": { "name": "screenplay-structure-analysis", "summary": "Analyse de la structure dramatique : cartographie l'ossature actes/scènes (zhe/chu) et la progression de l'intrigue des drames classiques du domaine public. Les données sont interrogées en temps réel dans le corpus de pièces du moteur fd-text-core (préréglage : structure)." }, "ja": { "name": "screenplay-structure-analysis", "summary": "脚本構造分析：パブリックドメインの古典戯曲の幕場（折/齣）の骨格と筋書きの展開を整理します。データは fd-text-core エンジンの脚本コーパスからリアルタイムに検索します（プリセット：structure）。" } }, "whenToUse": "Relevant when working on: 文本引擎, 剧本, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:screenplay-character-search", "slug": "skill-screenplay-character-search", "category": "skill", "marketCategory": null, "tags": [ "文本引擎", "剧本", "finddata" ], "canonical": { "brief": "Script character search and analysis: looks up a character's lines, appearance passages, and characterization features (dialogue aggregation, linguistic features, relationship networks). Data via the fd-text-core engine (preset: character)." }, "human": { "zh-CN": { "name": "screenplay-character-search", "summary": "剧本角色检索与分析：查某角色的台词、出场段落与形象特征（台词聚合、语言特征、关系网络）。数据经 fd-text-core 引擎（预设 character）。" }, "en": { "name": "screenplay-character-search", "summary": "Script character search and analysis: looks up a character's lines, appearance passages, and characterization features (dialogue aggregation, linguistic features, relationship networks). Data via the fd-text-core engine (preset: character)." }, "es": { "name": "screenplay-character-search", "summary": "Búsqueda y análisis de personajes: consulta las réplicas, los pasajes de aparición y los rasgos de caracterización de un personaje (agregación de diálogos, rasgos lingüísticos, redes de relaciones). Datos vía el motor fd-text-core (preset: character)." }, "fr": { "name": "screenplay-character-search", "summary": "Recherche et analyse de personnages : retrouve les répliques, les passages d'apparition et les traits de caractérisation d'un personnage (agrégation des dialogues, traits linguistiques, réseaux de relations). Données via le moteur fd-text-core (préréglage : character)." }, "ja": { "name": "screenplay-character-search", "summary": "脚本キャラクターの検索と分析：あるキャラクターの台詞、登場箇所、人物像の特徴（台詞の集約、言語的特徴、関係ネットワーク）を調べます。データは fd-text-core エンジンを使用します（プリセット：character）。" } }, "whenToUse": "Relevant when working on: 文本引擎, 剧本, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:screenplay-scene-search", "slug": "skill-screenplay-scene-search", "category": "skill", "marketCategory": null, "tags": [ "文本引擎", "剧本", "finddata" ], "canonical": { "brief": "Script scene search and close reading: locates scenes by description and closely reads the full scene (scene function, staging features, dramatic tension). Data via the fd-text-core engine (preset: scene)." }, "human": { "zh-CN": { "name": "screenplay-scene-search", "summary": "剧本场景检索与细读：按场景描述定位并细读整场内容（场景功能、调度特征、戏剧张力）。数据经 fd-text-core 引擎（预设 scene）。" }, "en": { "name": "screenplay-scene-search", "summary": "Script scene search and close reading: locates scenes by description and closely reads the full scene (scene function, staging features, dramatic tension). Data via the fd-text-core engine (preset: scene)." }, "es": { "name": "screenplay-scene-search", "summary": "Búsqueda y lectura detallada de escenas: localiza escenas por descripción y lee en detalle la escena completa (función dramática, rasgos de puesta en escena, tensión dramática). Datos vía el motor fd-text-core (preset: scene)." }, "fr": { "name": "screenplay-scene-search", "summary": "Recherche et lecture approfondie de scènes : localise les scènes par description et lit en détail la scène entière (fonction de la scène, caractéristiques de mise en scène, tension dramatique). Données via le moteur fd-text-core (préréglage : scene)." }, "ja": { "name": "screenplay-scene-search", "summary": "脚本シーンの検索と精読：シーンの説明から該当箇所を特定し、シーン全体を精読します（シーンの機能、演出の特徴、ドラマの緊張）。データは fd-text-core エンジンを使用します（プリセット：scene）。" } }, "whenToUse": "Relevant when working on: 文本引擎, 剧本, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:screenplay-similar-passages", "slug": "skill-screenplay-similar-passages", "category": "skill", "marketCategory": null, "tags": [ "文本引擎", "剧本", "finddata" ], "canonical": { "brief": "Script similar-passage search and comparison: uses a reference text to find semantically similar passages elsewhere, for cross-play comparison and motif tracing. Data via the fd-text-core engine (preset: similar, pure vector channel)." }, "human": { "zh-CN": { "name": "screenplay-similar-passages", "summary": "剧本相似段落检索与比较：以参考文本找语义相近的其他段落，做跨剧比较与母题溯源。数据经 fd-text-core 引擎（预设 similar，纯向量通道）。" }, "en": { "name": "screenplay-similar-passages", "summary": "Script similar-passage search and comparison: uses a reference text to find semantically similar passages elsewhere, for cross-play comparison and motif tracing. Data via the fd-text-core engine (preset: similar, pure vector channel)." }, "es": { "name": "screenplay-similar-passages", "summary": "Búsqueda y comparación de pasajes similares: usa un texto de referencia para hallar otros pasajes semánticamente afines, para comparación entre obras y rastreo de motivos. Datos vía el motor fd-text-core (preset: similar, canal puramente vectorial)." }, "fr": { "name": "screenplay-similar-passages", "summary": "Recherche et comparaison de passages similaires : à partir d'un texte de référence, trouve d'autres passages sémantiquement proches, pour des comparaisons entre pièces et le traçage de motifs. Données via le moteur fd-text-core (préréglage : similar, canal purement vectoriel)." }, "ja": { "name": "screenplay-similar-passages", "summary": "脚本の類似段落検索と比較：参照テキストから意味的に近い他の段落を見つけ、作品横断の比較やモチーフの源泉をたどります。データは fd-text-core エンジンを使用します（プリセット：similar、純ベクトル経路）。" } }, "whenToUse": "Relevant when working on: 文本引擎, 剧本, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:paper-literature-search", "slug": "skill-paper-literature-search", "category": "skill", "marketCategory": null, "tags": [ "text", "finddata" ], "canonical": { "brief": "Academic literature search (text engine · papers pack entry skill). Use when the user wants to find papers on a research direction, build a literature review, learn what work exists in a field, or asks 'find papers about XX' or 'what is the state of research in this direction'. Drives the workflow f" }, "human": { "zh-CN": { "name": "paper-literature-search", "summary": "论文文献检索（文本引擎·论文包入口技能）。当用户想查某个研究方向的论文、做文献综述、了解某领域有哪些工作、或要求\"找找关于 XX 的论文\"\"这个方向的研究现状\"时使用。驱动从方向定位到文献梳理的流程，数据经 fd-text-core 引擎的论文语料与外部开放检索（arXiv/OpenAlex/S2）实时获取。" }, "en": { "name": "paper-literature-search", "summary": "Academic literature search (text engine · papers pack entry skill). Use when the user wants to find papers on a research direction, build a literature review, learn what work exists in a field, or asks 'find papers about XX' or 'what is the state of research in this direction'. Drives the workflow from direction scoping to literature mapping; data comes from the fd-text-core engine's paper corpus plus external open search (arXiv/OpenAlex/S2), fetched in real time." }, "es": { "name": "paper-literature-search", "summary": "Búsqueda de literatura académica (skill de entrada del paquete de papers del motor de texto). Úselo cuando el usuario quiera buscar artículos sobre una línea de investigación, hacer una revisión de literatura, saber qué trabajos existen en un área, o pida «busca papers sobre XX» o «¿cuál es el estado de la investigación en esta dirección?». Dirige el flujo desde la delimitación de la dirección hasta el mapeo de la literatura; los datos provienen del corpus de papers del motor fd-text-core más búsqueda abierta externa (arXiv/OpenAlex/S2), obtenidos en tiempo real." }, "fr": { "name": "paper-literature-search", "summary": "Recherche de littérature académique (skill d'entrée du pack papers du moteur de texte). À utiliser lorsque l'utilisateur veut trouver des articles sur une direction de recherche, réaliser une revue de littérature, savoir quels travaux existent dans un domaine, ou demande « trouve des papiers sur XX » ou « quel est l'état de la recherche dans cette direction ». Pilote le flux depuis le cadrage de la direction jusqu'à la cartographie de la littérature ; les données proviennent du corpus de papiers du moteur fd-text-core plus la recherche ouverte externe (arXiv/OpenAlex/S2), en temps réel." }, "ja": { "name": "paper-literature-search", "summary": "論文文献検索（テキストエンジン・論文パックの入口スキル）。ある研究方向の論文を調べたい、文献レビューを作りたい、ある分野にどんな研究があるか知りたい、「XX についての論文を探して」「この方向の研究状況は？」などと求められたときに使用します。方向の特定から文献整理までのフローを駆動し、データは fd-text-core エンジンの論文コーパスと外部オープン検索（arXiv/OpenAlex/S2）からリアルタイムに取得します。" } }, "whenToUse": "Relevant when working on: text, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:paper-method-tracking", "slug": "skill-paper-method-tracking", "category": "skill", "marketCategory": null, "tags": [ "text", "finddata" ], "canonical": { "brief": "Paper method tracking (text engine · papers pack). Use when the user wants to know how a method/technique (e.g. transformer, contrastive learning, GARCH) is used in academic papers, what variants exist, or who proposed it, or asks to 'trace the development of the XX method'. Drives retrieval and lin" }, "human": { "zh-CN": { "name": "paper-method-tracking", "summary": "论文方法追踪（文本引擎·论文包）。当用户想查某个方法/技术（如 transformer、对比学习、GARCH）在学术论文里怎么用、有哪些变体、谁提出的，或要求\"追一下 XX 方法的发展\"时使用。按方法名驱动检索与脉络梳理，数据经 fd-text-core 引擎实时取。" }, "en": { "name": "paper-method-tracking", "summary": "Paper method tracking (text engine · papers pack). Use when the user wants to know how a method/technique (e.g. transformer, contrastive learning, GARCH) is used in academic papers, what variants exist, or who proposed it, or asks to 'trace the development of the XX method'. Drives retrieval and lineage mapping by method name; data fetched in real time via the fd-text-core engine." }, "es": { "name": "paper-method-tracking", "summary": "Seguimiento de métodos en papers (motor de texto · paquete de papers). Úselo cuando el usuario quiera saber cómo se usa un método/técnica (p. ej. transformer, aprendizaje contrastivo, GARCH) en artículos académicos, qué variantes existen o quién lo propuso, o pida «trazar la evolución del método XX». Dirige la búsqueda y el trazado del linaje por nombre de método; datos obtenidos en tiempo real vía el motor fd-text-core." }, "fr": { "name": "paper-method-tracking", "summary": "Suivi des méthodes dans les papiers (moteur de texte · pack papers). À utiliser lorsque l'utilisateur veut savoir comment une méthode/technique (p. ex. transformer, apprentissage contrastif, GARCH) est utilisée dans les articles académiques, quelles variantes existent ou qui l'a proposée, ou demande de « retracer l'évolution de la méthode XX ». Pilote la recherche et la cartographie de la filiation par nom de méthode ; données obtenues en temps réel via le moteur fd-text-core." }, "ja": { "name": "paper-method-tracking", "summary": "論文メソッドの追跡（テキストエンジン・論文パック）。ある手法/技術（transformer、対比学習、GARCH など）が学術論文でどう使われているか、どんな派生形があるか、誰が提唱したかを調べたい、「XX 手法の発展を追って」と求められたときに使用します。手法名を軸に検索と系譜の整理を駆動し、データは fd-text-core エンジンからリアルタイムに取得します。" } }, "whenToUse": "Relevant when working on: text, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:paper-related-work", "slug": "skill-paper-related-work", "category": "skill", "marketCategory": null, "tags": [ "text", "finddata" ], "canonical": { "brief": "Paper related-work and similar-literature search (text engine · papers pack). Use when the user provides an abstract/title/research idea and asks to find similar papers, write a related-work paragraph, or run a duplication-check-style comparison ('has anything similar been done?'). Finds nearest nei" }, "human": { "zh-CN": { "name": "paper-related-work", "summary": "论文相关工作与相似文献（文本引擎·论文包）。当用户给出一段摘要/题目/研究想法，要求找相似论文、写相关工作段落、或做查重式比对（\"有没有做过类似的\"）时使用。按参考文本的语义向量找近邻，数据经 fd-text-core 引擎实时取。" }, "en": { "name": "paper-related-work", "summary": "Paper related-work and similar-literature search (text engine · papers pack). Use when the user provides an abstract/title/research idea and asks to find similar papers, write a related-work paragraph, or run a duplication-check-style comparison ('has anything similar been done?'). Finds nearest neighbors by semantic vectors of the reference text; data fetched in real time via the fd-text-core engine." }, "es": { "name": "paper-related-work", "summary": "Trabajos relacionados y literatura similar (motor de texto · paquete de papers). Úselo cuando el usuario aporta un resumen/título/idea de investigación y pide encontrar artículos similares, escribir un párrafo de trabajos relacionados o hacer una comparación tipo antiplagio («¿ya se ha hecho algo similar?»). Encuentra los vecinos más cercanos por vectores semánticos del texto de referencia; datos obtenidos en tiempo real vía el motor fd-text-core." }, "fr": { "name": "paper-related-work", "summary": "Travaux liés et littérature similaire (moteur de texte · pack papers). À utiliser lorsque l'utilisateur fournit un résumé/titre/idée de recherche et demande de trouver des articles similaires, rédiger un paragraphe d'état de l'art, ou faire une comparaison type anti-plagiat (« est-ce que quelque chose de similaire a déjà été fait ? »). Trouve les plus proches voisins par vecteurs sémantiques du texte de référence ; données obtenues en temps réel via le moteur fd-text-core." }, "ja": { "name": "paper-related-work", "summary": "論文の関連研究・類似文献（テキストエンジン・論文パック）。アブストラクト/タイトル/研究アイデアが与えられ、類似論文の探索、関連研究セクションの執筆、類似チェック（「似た研究はあるか」）を求められたときに使用します。参照テキストの意味ベクトルで近傍を検索し、データは fd-text-core エンジンからリアルタイムに取得します。" } }, "whenToUse": "Relevant when working on: text, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:wx-wizard-check", "slug": "skill-wx-wizard-check", "category": "skill", "marketCategory": null, "tags": [], "canonical": { "brief": "Acceptance placeholder skill." }, "human": { "zh-CN": { "name": "wx-wizard-check", "summary": "验收占位技能" }, "en": { "name": "wx-wizard-check", "summary": "Acceptance placeholder skill." }, "es": { "name": "wx-wizard-check", "summary": "Skill marcador de posición para la aceptación." }, "fr": { "name": "wx-wizard-check", "summary": "Skill espace réservé pour la recette." }, "ja": { "name": "wx-wizard-check", "summary": "受け入れ検証用のプレースホルダースキル。" } }, "whenToUse": "验收占位技能", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:quick-look", "slug": "skill-quick-look", "category": "skill", "marketCategory": null, "tags": [], "canonical": { "brief": "Quick-look placeholder." }, "human": { "zh-CN": { "name": "quick-look", "summary": "速览占位" }, "en": { "name": "quick-look", "summary": "Quick-look placeholder." }, "es": { "name": "quick-look", "summary": "Marcador de posición de vista rápida." }, "fr": { "name": "quick-look", "summary": "Espace réservé d'aperçu rapide." }, "ja": { "name": "quick-look", "summary": "クイックルックのプレースホルダー。" } }, "whenToUse": "速览占位", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:brief-draft", "slug": "skill-brief-draft", "category": "skill", "marketCategory": null, "tags": [], "canonical": { "brief": "Brief placeholder." }, "human": { "zh-CN": { "name": "brief-draft", "summary": "简报占位" }, "en": { "name": "brief-draft", "summary": "Brief placeholder." }, "es": { "name": "brief-draft", "summary": "Marcador de posición de nota breve." }, "fr": { "name": "brief-draft", "summary": "Espace réservé de brief." }, "ja": { "name": "brief-draft", "summary": "ブリーフのプレースホルダー。" } }, "whenToUse": "简报占位", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:macro-cycle-analysis", "slug": "skill-macro-cycle-analysis", "category": "skill", "marketCategory": null, "tags": [ "wire", "经济", "宏观", "finddata" ], "canonical": { "brief": "Macro cycle and business-climate analysis. Use when the user asks to locate the current position in the economic cycle (recovery/overheating/stagflation/recession), run a six-domain climate scan (growth/inflation/money and credit/fiscal/foreign trade/real estate and consumption), layer leading-coinc" }, "human": { "zh-CN": { "name": "macro-cycle-analysis", "summary": "宏观周期与景气分析。当用户要求判断当前经济周期位置（复苏/过热/滞胀/衰退）、做六域景气扫描（增长/通胀/货币信用/财政/外贸/地产消费）、做领先-同步-滞后指标分层、识别景气拐点、梳理宏观变量到其业务问题的传导链，或做省际补充与国际对比时使用。走柏讯服务端宏观数据，不含收益率曲线与互换分析。" }, "en": { "name": "macro-cycle-analysis", "summary": "Macro cycle and business-climate analysis. Use when the user asks to locate the current position in the economic cycle (recovery/overheating/stagflation/recession), run a six-domain climate scan (growth/inflation/money and credit/fiscal/foreign trade/real estate and consumption), layer leading-coincident-lagging indicators, identify climate turning points, map transmission chains from macro variables to their business question, or do cross-province supplements and international comparisons. Uses 柏讯 server-side macro data; excludes yield-curve and swap analysis." }, "es": { "name": "macro-cycle-analysis", "summary": "Análisis del ciclo macro y del clima de negocios. Úselo cuando el usuario pida situar la posición actual en el ciclo económico (recuperación/sobrecalentamiento/estanflación/recesión), hacer un barrido del clima en seis dominios (crecimiento/inflación/moneda y crédito/fiscal/comercio exterior/inmobiliaria y consumo), estratificar indicadores adelantados-coincidentes-retrasados, identificar puntos de inflexión del ciclo, trazar cadenas de transmisión de variables macro a su problema de negocio, o añadir comparaciones interprovinciales e internacionales. Usa datos macro del servidor de 柏讯; excluye el análisis de curva de rendimientos y swaps." }, "fr": { "name": "macro-cycle-analysis", "summary": "Analyse du cycle macro et du climat des affaires. À utiliser lorsque l'utilisateur demande de situer la position actuelle dans le cycle économique (reprise/surchauffe/stagflation/récession), réaliser un balayage du climat sur six domaines (croissance/inflation/monnaie et crédit/budget/commerce extérieur/immobilier et consommation), stratifier les indicateurs avancés-coïncidants-retardés, identifier les points de retournement, cartographier les chaînes de transmission des variables macro vers sa question métier, ou compléter par des comparaisons interprovinciales et internationales. Utilise les données macro côté serveur de 柏讯 ; exclut l'analyse de courbe de taux et de swaps." }, "ja": { "name": "macro-cycle-analysis", "summary": "マクロ循環と景気分析。現在の経済循環の位置（回復/過熱/スタグフレーション/後退）の判断、六領域の景気スキャン（成長/インフレ/通貨・信用/財政/対外貿易/不動産・消費）、先行・一致・遅行指標の層別、景気転換点の特定、マクロ変数から自社の業務課題への伝達経路の整理、省際の補足や国際比較を求められたときに使用します。柏讯サーバー側のマクロデータを使用し、利回り曲線とスワップ分析は含みません。" } }, "whenToUse": "Relevant when working on: wire, 经济, 宏观, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/wire-presets/tree/main/skills/macro-cycle-analysis" }, { "id": "skill:industry-chain-analysis", "slug": "skill-industry-chain-analysis", "category": "skill", "marketCategory": null, "tags": [ "wire", "经济", "行业", "finddata" ], "canonical": { "brief": "Industry climate and value-chain transmission analysis. Use when the user wants to determine where an industry sits in the business cycle, map upstream/downstream input-output and cost-price transmission paths, substantiate industry judgments with listed-company financial panels, characterize the in" }, "human": { "zh-CN": { "name": "industry-chain-analysis", "summary": "行业景气与产业链传导分析。当用户要判断一个行业处在景气周期的哪个位置、梳理上下游投入产出与成本价格传导路径、用上市公司财务面板实证行业判断、刻画行业格局（规模/增速/集中度/进入壁垒），或把研报观点与公司真实财务数据对照时使用。走柏讯服务端行业与公司财务数据，不涉及价格与市值分析。" }, "en": { "name": "industry-chain-analysis", "summary": "Industry climate and value-chain transmission analysis. Use when the user wants to determine where an industry sits in the business cycle, map upstream/downstream input-output and cost-price transmission paths, substantiate industry judgments with listed-company financial panels, characterize the industry landscape (size/growth/concentration/entry barriers), or check research-report views against companies' actual financial data. Uses 柏讯 server-side industry and company financial data; does not cover price or market-cap analysis." }, "es": { "name": "industry-chain-analysis", "summary": "Análisis del clima sectorial y de la transmisión en la cadena industrial. Úselo cuando el usuario quiera determinar en qué punto del ciclo está una industria, trazar las rutas de transmisión de insumos-productos y costos-precios aguas arriba/abajo, validar juicios sectoriales con paneles financieros de empresas cotizadas, caracterizar la estructura de la industria (tamaño/crecimiento/concentración/barreras de entrada), o contrastar las opiniones de informes de investigación con los datos financieros reales de las empresas. Usa datos sectoriales y financieros del servidor de 柏讯; no cubre el análisis de precios ni de capitalización." }, "fr": { "name": "industry-chain-analysis", "summary": "Analyse du climat sectoriel et de la transmission le long de la chaîne industrielle. À utiliser lorsque l'utilisateur veut situer une industrie dans le cycle, cartographier les chemins de transmission amont/aval des intrants et des coûts-prix, étayer les jugements sectoriels par les panneaux financiers des sociétés cotées, caractériser la structure de l'industrie (taille/croissance/concentration/barrières à l'entrée), ou confronter les vues des rapports de recherche aux données financières réelles des entreprises. Utilise les données sectorielles et financières côté serveur de 柏讯 ; ne couvre pas l'analyse des prix ni de la capitalisation." }, "ja": { "name": "industry-chain-analysis", "summary": "業種景気と産業チェーン伝達の分析。ある業種が景気循環のどの位置にあるかの判断、上流・下流の投入産出とコスト・価格の伝達経路の整理、上場企業の財務パネルによる業種判断の実証、業界構造（規模/成長率/集中度/参入障壁）の描画、または調査レポートの見解と企業の実財務データの突き合わせを求められたときに使用します。柏讯サーバー側の業種・企業財務データを使用し、価格と時価総額の分析は行いません。" } }, "whenToUse": "Relevant when working on: wire, 经济, 行业, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/wire-presets/tree/main/skills/industry-chain-analysis" }, { "id": "skill:regional-economy-compare", "slug": "skill-regional-economy-compare", "category": "skill", "marketCategory": null, "tags": [ "wire", "经济", "区域", "finddata" ], "canonical": { "brief": "Regional economic comparison: multi-dimensional indicator rankings, percentile positioning, growth-structure decomposition, and similar-city clustering for Chinese cities/provinces. Use when the user wants to compare the economic strength, industrial structure, fiscal capacity, and household income " }, "human": { "zh-CN": { "name": "regional-economy-compare", "summary": "区域经济比较：对中国城市/省份做多维指标横向排名、分位定位、增长结构分解与相似城市聚类。当用户要比较两个或多个城市/省份的经济实力、产业结构、财政与居民收入，要某城市在全省或全国的排名与分位，要拆解经济总量增长的产业贡献与拉动点数，或要按结构找与目标城市可比的对照组时使用。数据走柏讯城市面板与统计年鉴。" }, "en": { "name": "regional-economy-compare", "summary": "Regional economic comparison: multi-dimensional indicator rankings, percentile positioning, growth-structure decomposition, and similar-city clustering for Chinese cities/provinces. Use when the user wants to compare the economic strength, industrial structure, fiscal capacity, and household income of two or more cities/provinces; wants a city's rank and percentile within its province or nationwide; wants the industrial contributions and pull points behind total growth decomposed; or wants a structurally comparable control group for a target city. Data comes from 柏讯 city panels and statistical yearbooks." }, "es": { "name": "regional-economy-compare", "summary": "Comparación económica regional: rankings multidimensionales de indicadores, posicionamiento por percentiles, descomposición de la estructura de crecimiento y agrupamiento de ciudades similares para ciudades/provincias de China. Úselo cuando el usuario quiera comparar la fortaleza económica, la estructura industrial, la hacienda y los ingresos de los hogares de dos o más ciudades/provincias; conocer el ranking y el percentil de una ciudad en su provincia o en el país; descomponer las contribuciones sectoriales y los puntos de arrastre del crecimiento del total; o buscar un grupo de control estructuralmente comparable para una ciudad objetivo. Datos de los paneles urbanos y los anuarios estadísticos de 柏讯." }, "fr": { "name": "regional-economy-compare", "summary": "Comparaison économique régionale : classements multi-indicateurs, positionnement en percentiles, décomposition de la structure de croissance et regroupement de villes similaires pour les villes/provinces de Chine. À utiliser lorsque l'utilisateur veut comparer la puissance économique, la structure industrielle, les finances et le revenu des ménages de deux villes/provinces ou plus ; connaître le rang et le percentile d'une ville dans sa province ou dans le pays ; décomposer les contributions sectorielles et les points d'entraînement de la croissance du total ; ou trouver un groupe témoin structurellement comparable pour une ville cible. Données issues des panneaux urbains et des annuaires statistiques de 柏讯." }, "ja": { "name": "regional-economy-compare", "summary": "地域経済の比較：中国の都市/省について多次元指標の横断ランキング、パーセンタイル位置づけ、成長構造の分解、類似都市のクラスタリングを行います。2つ以上の都市/省の経済力、産業構造、財政、住民所得の比較、ある都市の省内または全国での順位とパーセンタイル、経済総量の成長に対する産業別寄与と牽引ポイントの分解、または構造から見て対象都市と比較可能な対照群の探索を求められたときに使用します。データは柏讯の都市パネルと統計年鑑を使用します。" } }, "whenToUse": "Relevant when working on: wire, 经济, 区域, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/wire-presets/tree/main/skills/regional-economy-compare" }, { "id": "skill:corporate-fundamental-analysis", "slug": "skill-corporate-fundamental-analysis", "category": "skill", "marketCategory": null, "tags": [ "wire", "金融", "公司", "finddata" ], "canonical": { "brief": "Corporate fundamentals and financial quality analysis (柏讯 line). Use when the user asks for a quick three-statement review, DuPont decomposition, cash-flow quality diagnostics, financial red-flag screening, or peer percentile comparison for an A-share listed company. Draws on GTA firm-year financial" }, "human": { "zh-CN": { "name": "corporate-fundamental-analysis", "summary": "企业基本面与财务质量分析（柏讯线）。当用户要求给一家 A 股上市公司做三表速览、杜邦分解、现金流质量诊断、财务红旗项排查或同业分位对照时使用。走 GTA firm-year 财务面板 + 中国财报三表与附注，输出带来源与数据期数的可复核结论。本技能只做基本面质量判断，不做估值、不给目标价。" }, "en": { "name": "corporate-fundamental-analysis", "summary": "Corporate fundamentals and financial quality analysis (柏讯 line). Use when the user asks for a quick three-statement review, DuPont decomposition, cash-flow quality diagnostics, financial red-flag screening, or peer percentile comparison for an A-share listed company. Draws on GTA firm-year financial panels plus the three main statements and notes of Chinese financial reports, producing verifiable conclusions with sources and data periods. This skill only assesses fundamental quality — no valuation, no target prices." }, "es": { "name": "corporate-fundamental-analysis", "summary": "Análisis de fundamentales y calidad financiera (línea 柏讯). Úselo cuando el usuario pida una revisión rápida de los tres estados, descomposición DuPont, diagnóstico de calidad del flujo de caja, cribado de banderas rojas financieras o comparación de percentiles con pares para una empresa cotizada en A-shares. Se apoya en paneles financieros firm-year de GTA más los tres estados y las notas de los informes financieros chinos, y produce conclusiones verificables con fuentes y periodos de datos. Esta skill solo juzga la calidad fundamental: sin valoración y sin precios objetivo." }, "fr": { "name": "corporate-fundamental-analysis", "summary": "Analyse fondamentale et qualité financière (ligne 柏讯). À utiliser lorsque l'utilisateur demande une revue rapide des trois états, une décomposition DuPont, un diagnostic de la qualité du cash-flow, un criblage des signaux d'alerte financiers ou une comparaison en percentiles avec les pairs pour une société cotée A-share. S'appuie sur les panneaux financiers firm-year de GTA plus les trois états et les annexes des rapports financiers chinois, et produit des conclusions vérifiables avec sources et périodes de données. Cette skill ne juge que la qualité fondamentale : pas de valorisation, pas de prix cible." }, "ja": { "name": "corporate-fundamental-analysis", "summary": "企業ファンダメンタルズと財務品質の分析（柏讯ライン）。A 股上場企業について三表のクイックレビュー、デュポン分解、キャッシュフロー品質の診断、財務レッドフラグの点検、同業他社とのパーセンタイル比較を求められたときに使用します。GTA の firm-year 財務パネルと中国財報の三表・附注を使用し、出典とデータ期数を添えた検証可能な結論を出力します。本スキルはファンダメンタルズの品質判定のみを行い、バリュエーションや目標株価は出しません。" } }, "whenToUse": "Relevant when working on: wire, 金融, 公司, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/wire-presets/tree/main/skills/corporate-fundamental-analysis" }, { "id": "skill:panel-econometrics", "slug": "skill-panel-econometrics", "category": "skill", "marketCategory": null, "tags": [ "wire", "经济", "计量", "finddata" ], "canonical": { "brief": "Panel data econometrics. Use when the user wants inference on a two-dimensional 'unit × time' structure — fixed/random effects with the Hausman test, cluster-robust standard errors, dynamic panels (lagged dependent variables), difference-in-differences DID with parallel-trends/event-study tests, and" }, "human": { "zh-CN": { "name": "panel-econometrics", "summary": "面板数据计量分析。当用户要在\"个体 × 时间\"二维结构上做推断时使用——固定效应/随机效应与 Hausman 检验、聚类稳健标准误、动态面板（滞后因变量）、双重差分 DID 与平行趋势/事件研究检验、断点回归概念。面板取自柏讯服务端三类源：GTA 上市公司（firm-year）、中国城市（city-year）、世界银行 WDI（country-year），宏观控制变量取柏讯宏观序列。不涉及股价与市值建模。" }, "en": { "name": "panel-econometrics", "summary": "Panel data econometrics. Use when the user wants inference on a two-dimensional 'unit × time' structure — fixed/random effects with the Hausman test, cluster-robust standard errors, dynamic panels (lagged dependent variables), difference-in-differences DID with parallel-trends/event-study tests, and regression discontinuity concepts. Panels come from three 柏讯 server-side sources: GTA listed companies (firm-year), Chinese cities (city-year), and World Bank WDI (country-year); macro control variables come from 柏讯 macro series. Does not cover stock price or market-cap modeling." }, "es": { "name": "panel-econometrics", "summary": "Econometría de datos de panel. Úselo cuando el usuario quiera inferencia sobre una estructura bidimensional 'unidad × tiempo' — efectos fijos/aleatorios con prueba de Hausman, errores estándar robustos por clúster, paneles dinámicos (variable dependiente rezagada), diferencias-en-diferencias DID con pruebas de tendencias paralelas/eventos, y conceptos de regresión discontinua. Los paneles provienen de tres fuentes del servidor de 柏讯: empresas cotizadas de GTA (firm-year), ciudades chinas (city-year) y WDI del Banco Mundial (country-year); las variables de control macro provienen de las series macro de 柏讯. No cubre el modelado de precios de acciones ni de capitalización." }, "fr": { "name": "panel-econometrics", "summary": "Économétrie des données de panel. À utiliser lorsque l'utilisateur veut faire de l'inférence sur une structure bidimensionnelle « unité × temps » — effets fixes/aléatoires avec test de Hausman, erreurs standards robustes par cluster, panels dynamiques (variable dépendante retardée), différences-de-différences DID avec tests de tendances parallèles/event study, et concepts de régression sur discontinuité. Les panels proviennent de trois sources côté serveur de 柏讯 : sociétés cotées GTA (firm-year), villes chinoises (city-year) et WDI de la Banque mondiale (country-year) ; les variables de contrôle macro proviennent des séries macro de 柏讯. Ne couvre pas la modélisation des cours ni de la capitalisation." }, "ja": { "name": "panel-econometrics", "summary": "パネルデータの計量分析。「個体 × 時間」の二次元構造での推測を求められたときに使用します — 固定効果/ランダム効果と Hausman 検定、クラスターロバスト標準誤差、動的パネル（ラグ付き従属変数）、差分の差分 DID とパラレルトレンド/イベントスタディ検定、回帰不連続の概念。パネルは柏讯サーバー側の3種のソースから取得：GTA 上場企業（firm-year）、中国都市（city-year）、世界銀行 WDI（country-year）。マクロ制御変数は柏讯のマクロ系列を使用。株価と時価総額のモデリングは行いません。" } }, "whenToUse": "Relevant when working on: wire, 经济, 计量, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/wire-presets/tree/main/skills/panel-econometrics" }, { "id": "skill:statistical-inference", "slug": "skill-statistical-inference", "category": "skill", "marketCategory": null, "tags": [ "wire", "经济", "统计", "finddata" ], "canonical": { "brief": "Statistical testing and exploratory analysis. Use when the user asks for descriptive statistics and distribution diagnostics, hypothesis test selection (t/ANOVA/chi-square/non-parametric), significance testing of differences between two groups or cities, correlation and paired analysis, time-series " }, "human": { "zh-CN": { "name": "statistical-inference", "summary": "统计检验与探索性分析。当用户要求做描述统计与分布诊断、选假设检验（t/ANOVA/卡方/非参）、判断两组或城市间差异是否显著、做相关性与配对分析、诊断时序平稳性与单位根、做多重比较校正，或对已有的本地 CSV/表格做探索性分析时使用。走柏讯服务端序列取数，价格类序列走本地 DAAS。" }, "en": { "name": "statistical-inference", "summary": "Statistical testing and exploratory analysis. Use when the user asks for descriptive statistics and distribution diagnostics, hypothesis test selection (t/ANOVA/chi-square/non-parametric), significance testing of differences between two groups or cities, correlation and paired analysis, time-series stationarity and unit-root diagnostics, multiple-comparison correction, or exploratory analysis of existing local CSV/tables. Series data is fetched via the 柏讯 server side; price-type series use local DAAS." }, "es": { "name": "statistical-inference", "summary": "Pruebas estadísticas y análisis exploratorio. Úselo cuando el usuario pida estadística descriptiva y diagnóstico de distribuciones, selección de pruebas de hipótesis (t/ANOVA/chi-cuadrado/no paramétricas), significancia de las diferencias entre dos grupos o ciudades, análisis de correlación y pareado, diagnóstico de estacionariedad y raíces unitarias en series temporales, corrección por comparaciones múltiples, o análisis exploratorio de CSV/tablas locales existentes. Las series se obtienen vía el servidor de 柏讯; las series de tipo precio usan DAAS local." }, "fr": { "name": "statistical-inference", "summary": "Tests statistiques et analyse exploratoire. À utiliser lorsque l'utilisateur demande des statistiques descriptives et des diagnostics de distribution, le choix de tests d'hypothèses (t/ANOVA/chi-deux/non paramétriques), la significativité des écarts entre deux groupes ou villes, l'analyse de corrélations et d'appariements, les diagnostics de stationnarité et de racine unitaire des séries temporelles, la correction des comparaisons multiples, ou l'analyse exploratoire de CSV/tableaux locaux existants. Les séries sont récupérées côté serveur 柏讯 ; les séries de type prix utilisent le DAAS local." }, "ja": { "name": "statistical-inference", "summary": "統計検定と探索的データ分析。記述統計と分布の診断、仮説検定の選択（t/ANOVA/カイ二乗/ノンパラメトリック）、2群間または都市間の差の有意性の判断、相関とペア分析、時系列の定常性と単位根の診断、多重比較の補正、または既存のローカル CSV/表の探索的分析を求められたときに使用します。系列データは柏讯サーバー側から取得し、価格系の系列はローカル DAAS を使用します。" } }, "whenToUse": "Relevant when working on: wire, 经济, 统计, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/wire-presets/tree/main/skills/statistical-inference" }, { "id": "skill:equity-valuation", "slug": "skill-equity-valuation", "category": "skill", "marketCategory": null, "tags": [ "wire", "金融", "估值", "finddata" ], "canonical": { "brief": "Equity valuation (DCF + comparable companies + three-statement linkage). Use when the user asks for intrinsic or relative valuation of an A-share/HK-listed company, WACC and terminal value calculation, sensitivity and scenario tables, a target price range, peer multiple comparison, or back-solving t" }, "human": { "zh-CN": { "name": "equity-valuation", "summary": "股票估值（DCF + 可比公司 + 三表联动）。当用户要求给某家 A 股/港股公司做内在价值或相对估值、算 WACC 与终值、跑敏感性与情景表、给目标价区间、做同业倍数对照与反推市场隐含预期时使用。财务数据走柏讯服务端通道 A，价格与市值走本地 DAAS 通道 B；通道 B 未连接时中止估值，不给目标价。" }, "en": { "name": "equity-valuation", "summary": "Equity valuation (DCF + comparable companies + three-statement linkage). Use when the user asks for intrinsic or relative valuation of an A-share/HK-listed company, WACC and terminal value calculation, sensitivity and scenario tables, a target price range, peer multiple comparison, or back-solving the market's implied expectations. Financial data goes through 柏讯 server-side channel A; prices and market caps through local DAAS channel B; if channel B is not connected, the valuation halts and no target price is given." }, "es": { "name": "equity-valuation", "summary": "Valoración de acciones (DCF + comparables + vinculación de los tres estados). Úselo cuando el usuario pida valoración intrínseca o relativa de una empresa cotizada en A-shares/Hong Kong, cálculo de WACC y valor terminal, tablas de sensibilidad y escenarios, un rango de precio objetivo, comparación de múltiplos con pares o deducción inversa de las expectativas implícitas del mercado. Los datos financieros van por el canal A del servidor de 柏讯; precios y capitalización por el canal B de DAAS local; si el canal B no está conectado, la valoración se aborta y no se da precio objetivo." }, "fr": { "name": "equity-valuation", "summary": "Valorisation d'actions (DCF + sociétés comparables + articulation des trois états). À utiliser lorsque l'utilisateur demande une valorisation intrinsèque ou relative d'une société cotée A-share/Hong Kong, le calcul du WACC et de la valeur terminale, des tables de sensibilité et de scénarios, une fourchette de prix cible, la comparaison de multiples avec les pairs ou la rétro-ingénierie des attentes implicites du marché. Les données financières passent par le canal A côté serveur de 柏讯 ; prix et capitalisations par le canal B du DAAS local ; si le canal B n'est pas connecté, la valorisation s'interrompt et aucun prix cible n'est donné." }, "ja": { "name": "equity-valuation", "summary": "株価バリュエーション（DCF + 類似会社 + 三表連動）。A 股/香港上場企業の内在価値または相対バリュエーション、WACC とターミナルバリューの計算、感応度・シナリオ表、目標株価レンジ、同業他社のマルチプル比較、市場が織り込む期待の逆算を求められたときに使用します。財務データは柏讯サーバー側のチャネル A、価格と時価総額はローカル DAAS のチャネル B を使用します。チャネル B が未接続の場合はバリュエーションを中止し、目標株価は出しません。" } }, "whenToUse": "Relevant when working on: wire, 金融, 估值, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/wire-presets/tree/main/skills/equity-valuation" }, { "id": "skill:earnings-cycle-analysis", "slug": "skill-earnings-cycle-analysis", "category": "skill", "marketCategory": null, "tags": [ "wire", "金融", "财报", "finddata" ], "canonical": { "brief": "Earnings cycle analysis (preview + recap). Use when the user asks for an earnings preview for a company (type and timing of the coming disclosure, which line items to watch, where beats/misses may come from), an earnings recap (actuals vs. the same period last year, which line items drove the differ" }, "human": { "zh-CN": { "name": "earnings-cycle-analysis", "summary": "财报周期分析（前瞻 + 复盘）。当用户要求对一家公司做财报前瞻（本次披露类型与时间、该盯哪些科目、超预期/低于预期可能从哪来）、做财报复盘（实际 vs 上年同期、差异由哪个科目驱动、盈利质量如何、对后续季度意味着什么），或梳理连续数期经营动能变化时使用。基于柏讯财报与公告数据，不做技术分析。触发词：财报前瞻、业绩预告、财报复盘、业绩点评、超预期、低于预期、盈利质量、某某公司季报/半年报/年报分析。" }, "en": { "name": "earnings-cycle-analysis", "summary": "Earnings cycle analysis (preview + recap). Use when the user asks for an earnings preview for a company (type and timing of the coming disclosure, which line items to watch, where beats/misses may come from), an earnings recap (actuals vs. the same period last year, which line items drove the difference, how earnings quality stands, what it means for coming quarters), or a mapping of operating momentum changes over several consecutive periods. Based on 柏讯 earnings and announcement data; no technical analysis. Trigger words: earnings preview, results forecast, earnings recap, results commentary, beat, miss, earnings quality, and quarterly/interim/annual report analysis of a given company." }, "es": { "name": "earnings-cycle-analysis", "summary": "Análisis del ciclo de resultados (previsión + revisión). Úselo cuando el usuario pida una previsión de resultados de una empresa (tipo y momento de la próxima divulgación, qué partidas vigilar, de dónde pueden venir los desvíos al alza/a la baja), una revisión de resultados (real vs. mismo periodo del año anterior, qué partidas impulsaron la diferencia, cómo está la calidad de los beneficios, qué implica para los próximos trimestres), o el seguimiento de los cambios de impulso operativo a lo largo de varios periodos consecutivos. Basado en datos de resultados y anuncios de 柏讯; sin análisis técnico. Palabras de activación: previsión de resultados, avance de resultados, revisión de resultados, comentario de resultados, por encima/debajo de lo esperado, calidad de los beneficios, y análisis del informe trimestral/semestral/anual de una empresa." }, "fr": { "name": "earnings-cycle-analysis", "summary": "Analyse du cycle des résultats (anticipation + récapitulatif). À utiliser lorsque l'utilisateur demande une anticipation des résultats d'une société (type et date de la prochaine publication, quelles lignes surveiller, d'où peuvent venir les surprises hautes/basses), un récapitulatif des résultats (réalisé vs. même période de l'an dernier, quelles lignes ont piloté l'écart, quelle est la qualité des profits, ce que cela implique pour les prochains trimestres), ou une cartographie de l'évolution de la dynamique opérationnelle sur plusieurs périodes consécutives. Basé sur les données de résultats et d'annonces de 柏讯 ; pas d'analyse technique. Mots déclencheurs : anticipation des résultats, pré-annonce de résultats, récapitulatif des résultats, commentaire de résultats, au-dessus/en-dessous des attentes, qualité des profits, et analyse du rapport trimestriel/semestriel/annuel d'une société." }, "ja": { "name": "earnings-cycle-analysis", "summary": "決算サイクル分析（フォアキャスト + レビュー）。企業の決算フォアキャスト（今回の開示種別と時期、注目すべき科目、上振れ/下振れの要因候補）、決算レビュー（実績 vs 前年同期、差異を牽引した科目、収益品質、今後の四半期への含意）、または連続する複数期の事業モメンタムの変化の整理を求められたときに使用します。柏讯の決算・開示データに基づき、テクニカル分析は行いません。トリガーワード：決算フォアキャスト、業績予想、決算レビュー、業績コメント、サプライズ、未達、収益品質、特定企業の四半期/中間/年次報告分析。" } }, "whenToUse": "Relevant when working on: wire, 金融, 财报, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" }, "repositoryUrl": "https://gitee.com/FindDataTechnology/finddata-skills/tree/master/wire-presets/tree/main/skills/earnings-cycle-analysis" }, { "id": "skill:mb-business-checkup", "slug": "skill-mb-business-checkup", "category": "skill", "marketCategory": null, "tags": [ "micro-business", "个体户", "finddata" ], "canonical": { "brief": "Business checkup core entry (Facet · sole proprietors and micro-enterprises): four-stage diagnostic intake and profiling → single-store accounting → industry signal comparison → action list. All amounts and ratios are produced via calculators with formula and base source attached; if industry signal" }, "human": { "zh-CN": { "name": "mb-business-checkup", "summary": "生意体检内核入口（谦面·个体户与微型企业）：四段式问诊建档→单店算账→行业信号对照→行动清单。金额与比率一律经计算器产出并附公式与基数来源；未装行业信号包时如实标注不中断。" }, "en": { "name": "mb-business-checkup", "summary": "Business checkup core entry (Facet · sole proprietors and micro-enterprises): four-stage diagnostic intake and profiling → single-store accounting → industry signal comparison → action list. All amounts and ratios are produced via calculators with formula and base source attached; if industry signal packs are not installed, this is honestly noted without interruption." }, "es": { "name": "mb-business-checkup", "summary": "Entrada del núcleo de chequeo de negocio (Facet · autónomos y microempresas): admisión diagnóstica y creación de expediente en cuatro etapas → cuentas de la tienda única → contraste con señales del sector → lista de acciones. Importes y ratios se generan siempre vía calculadoras con fórmula y fuente de la base; si los paquetes de señales del sector no están instalados, se indica con honestidad sin interrumpir." }, "fr": { "name": "mb-business-checkup", "summary": "Entrée du noyau de diagnostic d'entreprise (Facet · indépendants et micro-entreprises) : entretien diagnostique et constitution du dossier en quatre étapes → comptes du magasin → confrontation aux signaux du secteur → liste d'actions. Montants et ratios sont toujours produits via des calculateurs avec formule et source de la base ; si les packs de signaux du secteur ne sont pas installés, cela est signalé honnêtement sans interrompre." }, "ja": { "name": "mb-business-checkup", "summary": "ビジネスヘルスチェックのコア入口（Facet · 個人事業主と零細企業向け）：四段階の問診・記録作成 → 単店の損益計算 → 業界シグナルとの照合 → アクションリスト。金額と比率はすべて計算機経由で出力し、公式と基数の出典を添付します。業界シグナルパックが未導入の場合は、中断せずその旨を正直に明記します。" } }, "whenToUse": "Relevant when working on: micro-business, 个体户, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:mb-breakeven-calculator", "slug": "skill-mb-breakeven-calculator", "category": "skill", "marketCategory": null, "tags": [ "micro-business", "个体户", "finddata" ], "canonical": { "brief": "Break-even calculator (Facet · sole proprietors): computes monthly break-even revenue, break-even customer count/order volume, safety margin, and days to break even. Every figure comes with formula + base + source; if an external base cannot be obtained, that item halts and is explicitly stated — ne" }, "human": { "zh-CN": { "name": "mb-breakeven-calculator", "summary": "盈亏平衡计算器（谦面·个体户）：算月保本流水、保本客流/单量、安全边际、保本天数。每个数字附公式+基数+来源；外部基数取不到即中止该项并明示，不用模型记忆数字填充。" }, "en": { "name": "mb-breakeven-calculator", "summary": "Break-even calculator (Facet · sole proprietors): computes monthly break-even revenue, break-even customer count/order volume, safety margin, and days to break even. Every figure comes with formula + base + source; if an external base cannot be obtained, that item halts and is explicitly stated — never filled with numbers from model memory." }, "es": { "name": "mb-breakeven-calculator", "summary": "Calculadora del punto de equilibrio (Facet · autónomos): calcula la facturación mensual de equilibrio, los clientes/pedidos de equilibrio, el margen de seguridad y los días hasta el equilibrio. Cada cifra lleva fórmula + base + fuente; si no se puede obtener una base externa, ese cálculo se detiene y se declara explícitamente, sin rellenarlo con cifras de la memoria del modelo." }, "fr": { "name": "mb-breakeven-calculator", "summary": "Calculateur de seuil de rentabilité (Facet · indépendants) : calcule le chiffre d'affaires mensuel au point mort, le nombre de clients/commandes au point mort, la marge de sécurité et les jours avant le point mort. Chaque chiffre comporte formule + base + source ; si une base externe est indisponible, le calcul s'arrête et est signalé explicitement — jamais comblé par des chiffres issus de la mémoire du modèle." }, "ja": { "name": "mb-breakeven-calculator", "summary": "損益分岐計算機（Facet · 個人事業主向け）：月次損益分岐売上、分岐に必要な客数/注文数、安全余裕、損益分岐までの日数を計算します。数値には必ず公式 + 基数 + 出典を添付します。外部基数が取得できない場合はその項目を中止して明示し、モデルの記憶にある数値で補填しません。" } }, "whenToUse": "Relevant when working on: micro-business, 个体户, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:mb-margin-structure-calculator", "slug": "skill-mb-margin-structure-calculator", "category": "skill", "marketCategory": null, "tags": [ "micro-business", "个体户", "finddata" ], "canonical": { "brief": "Gross margin structure calculator (Facet · sole proprietors): fixed/variable cost split, each cost as a share of revenue, gross margin and contribution margin, and cost-structure diagnostics. Every figure comes with formula + base + source; the split results feed the break-even calculator." }, "human": { "zh-CN": { "name": "mb-margin-structure-calculator", "summary": "毛利结构计算器（谦面·个体户）：固定/变动成本拆分、各项成本占流水比、毛利率与边际贡献率、成本结构诊断。每数字附公式+基数+来源，拆分结果供盈亏平衡计算器消费。" }, "en": { "name": "mb-margin-structure-calculator", "summary": "Gross margin structure calculator (Facet · sole proprietors): fixed/variable cost split, each cost as a share of revenue, gross margin and contribution margin, and cost-structure diagnostics. Every figure comes with formula + base + source; the split results feed the break-even calculator." }, "es": { "name": "mb-margin-structure-calculator", "summary": "Calculadora de estructura de margen bruto (Facet · autónomos): desglose de costos fijos/variables, cada costo como proporción de la facturación, margen bruto y margen de contribución, y diagnóstico de la estructura de costos. Cada cifra lleva fórmula + base + fuente; los resultados del desglose alimentan a la calculadora del punto de equilibrio." }, "fr": { "name": "mb-margin-structure-calculator", "summary": "Calculateur de structure de marge brute (Facet · indépendants) : ventilation coûts fixes/variables, poids de chaque coût dans le chiffre d'affaires, taux de marge brute et marge sur coûts variables, diagnostic de la structure des coûts. Chaque chiffre comporte formule + base + source ; les résultats de la ventilation alimentent le calculateur de seuil de rentabilité." }, "ja": { "name": "mb-margin-structure-calculator", "summary": "粗利構造計算機（Facet · 個人事業主向け）：固定費/変動費の分解、各コストの売上比率、粗利率と限界利益率、コスト構造の診断を行います。数値には必ず公式 + 基数 + 出典を添付し、分解結果は損益分岐計算機が利用します。" } }, "whenToUse": "Relevant when working on: micro-business, 个体户, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:mb-cashflow-terms-calculator", "slug": "skill-mb-cashflow-terms-calculator", "category": "skill", "marketCategory": null, "tags": [ "micro-business", "个体户", "finddata" ], "canonical": { "brief": "Cash flow and payment-terms calculator (Facet · sole proprietors): timing and size of cash-flow gaps, capital tied up in payment terms, minimum working capital, and seasonal buffers. Cash flow and profit/loss are presented separately; every figure comes with formula + base + source." }, "human": { "zh-CN": { "name": "mb-cashflow-terms-calculator", "summary": "现金流与账期计算器（谦面·个体户）：现金流缺口时点与金额、账期占款、最少周转金、季节性缓冲。现金流与盈亏分开呈现；每数字附公式+基数+来源。" }, "en": { "name": "mb-cashflow-terms-calculator", "summary": "Cash flow and payment-terms calculator (Facet · sole proprietors): timing and size of cash-flow gaps, capital tied up in payment terms, minimum working capital, and seasonal buffers. Cash flow and profit/loss are presented separately; every figure comes with formula + base + source." }, "es": { "name": "mb-cashflow-terms-calculator", "summary": "Calculadora de flujo de caja y plazos de cobro/pago (Facet · autónomos): momento y magnitud de los desfases de caja, capital inmovilizado en los plazos, mínimo de capital de trabajo y colchones estacionales. El flujo de caja y el resultado se presentan por separado; cada cifra lleva fórmula + base + fuente." }, "fr": { "name": "mb-cashflow-terms-calculator", "summary": "Calculateur de trésorerie et de délais de paiement (Facet · indépendants) : moment et ampleur des trous de trésorerie, capital immobilisé dans les délais, besoin minimal en fonds de roulement et amortisseurs saisonniers. Trésorerie et profit/perte sont présentés séparément ; chaque chiffre comporte formule + base + source." }, "ja": { "name": "mb-cashflow-terms-calculator", "summary": "キャッシュフローと支払サイトの計算機（Facet · 個人事業主向け）：資金繰りギャップの時点と金額、サイト資金の滞留、最低必要な運転資金、季節変動のバッファを計算します。キャッシュフローと損益は分けて提示し、数値には必ず公式 + 基数 + 出典を添付します。" } }, "whenToUse": "Relevant when working on: micro-business, 个体户, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:mb-signal-catering", "slug": "skill-mb-signal-catering", "category": "skill", "marketCategory": null, "tags": [ "micro-business", "餐饮", "finddata" ], "canonical": { "brief": "Catering industry signal pack (Facet · business checkup companion): food-cost trends (DCE futures), commercial-district activity proxy (专资办 box office), dine-in vs. delivery switching proxy (weather bureau), with optional peer density. Every signal is labeled with its proxy nature; business-registra" }, "human": { "zh-CN": { "name": "mb-signal-catering", "summary": "餐饮行业信号包（谦面·生意体检姊妹件）：食材成本趋势（大商所期货）、商圈活跃度代理（专资办票房）、堂食外卖切换代理（气象台），可选同业密度。每信号标注代理性质；工商类带批次年份与「历史格局非实时」，无注销口径不做关店量。" }, "en": { "name": "mb-signal-catering", "summary": "Catering industry signal pack (Facet · business checkup companion): food-cost trends (DCE futures), commercial-district activity proxy (专资办 box office), dine-in vs. delivery switching proxy (weather bureau), with optional peer density. Every signal is labeled with its proxy nature; business-registration signals carry their batch year and a 'historical pattern, not real-time' note; with no deregistration basis, no closure counts are produced." }, "es": { "name": "mb-signal-catering", "summary": "Paquete de señales del sector de restauración (Facet · complemento del chequeo de negocio): tendencias del costo de los alimentos (futuros de DCE), proxy de actividad de la zona comercial (taquilla de 专资办), proxy de la alternancia entre consumo en local y entrega (servicio meteorológico), y densidad de pares opcional. Cada señal se etiqueta con su carácter de proxy; las señales de registro empresarial llevan el año de lote y la nota «patrón histórico, no en tiempo real»; sin métrica de bajas registrales, no se producen recuentos de cierres." }, "fr": { "name": "mb-signal-catering", "summary": "Pack de signaux du secteur de la restauration (Facet · compagnon du diagnostic d'entreprise) : tendances du coût des denrées (futures DCE), proxy d'activité du quartier commercial (box-office 专资办), proxy de bascule sur place/livraison (service météo), densité concurrentielle optionnelle. Chaque signal est étiqueté selon sa nature de proxy ; les signaux d'immatriculation portent l'année de lot et la mention « structure historique, pas temps réel » ; sans statistique de radiation, aucun décompte de fermetures." }, "ja": { "name": "mb-signal-catering", "summary": "飲食業界シグナルパック（Facet · ビジネスチェックの姉妹件）：食材コストのトレンド（大商所先物）、商店街の賑わいの代理指標（专资办の興行成績）、店内食 vs デリバリー切替の代理指標（気象台）、同業密度（任意）。各シグナルには代理指標であることが明記され、工商登録系にはバッチ年と「歴史的な構造でありリアルタイムではない」の注記が付きます。注消統計がないため閉店数は出しません。" } }, "whenToUse": "Relevant when working on: micro-business, 餐饮, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:mb-signal-retail", "slug": "skill-mb-signal-retail", "category": "skill", "marketCategory": null, "tags": [ "micro-business", "零售", "finddata" ], "canonical": { "brief": "Retail industry signal pack (Facet · business checkup companion): wholesale price reference (Kunming flower auction), city fundamentals (city panel), peer density (business registrations, with sole-proprietorship filtering), and overall express-delivery sentiment. The business-registration data is t" }, "human": { "zh-CN": { "name": "mb-signal-retail", "summary": "零售行业信号包（谦面·生意体检姊妹件）：批发价参考（昆明花拍）、城市基本面（城市面板）、同业密度（工商登记，含个体工商户筛选）、快递大盘景气。工商数据批次 2023 滞后约 3 年，标「历史格局非实时」；无注销口径，关店量明示做不了。" }, "en": { "name": "mb-signal-retail", "summary": "Retail industry signal pack (Facet · business checkup companion): wholesale price reference (Kunming flower auction), city fundamentals (city panel), peer density (business registrations, with sole-proprietorship filtering), and overall express-delivery sentiment. The business-registration data is the 2023 batch, lagging about 3 years, marked 'historical pattern, not real-time'; with no deregistration basis, closure counts are explicitly stated as unavailable." }, "es": { "name": "mb-signal-retail", "summary": "Paquete de señales del sector minorista (Facet · complemento del chequeo de negocio): referencia de precios mayoristas (subasta de flores de Kunming), fundamentales de ciudad (panel urbano), densidad de pares (registros empresariales, con filtro de autónomos) y clima general del sector del paquete exprés. Los registros empresariales son del lote 2023, con unos 3 años de retraso, marcados «patrón histórico, no en tiempo real»; sin métrica de bajas registrales, se indica explícitamente que los recuentos de cierres no son posibles." }, "fr": { "name": "mb-signal-retail", "summary": "Pack de signaux du secteur du commerce de détail (Facet · compagnon du diagnostic d'entreprise) : référence de prix de gros (vente aux enchères de fleurs de Kunming), fondamentaux urbains (panneau de villes), densité concurrentielle (immatriculations, avec filtre des indépendants) et climat global de l'express. Les données d'immatriculation sont le lot 2023, avec environ 3 ans de retard, marquées « structure historique, pas temps réel » ; sans statistique de radiation, le décompte des fermetures est explicitement déclaré impossible." }, "ja": { "name": "mb-signal-retail", "summary": "小売業界シグナルパック（Facet · ビジネスチェックの姉妹件）：卸価格の参照（昆明の花オークション）、都市ファンダメンタルズ（都市パネル）、同業密度（工商登記、個人事業主のフィルタ付き）、宅配便全体の景況。工商データは2023年バッチで約3年の遅延があり、「歴史的な構造でありリアルタイムではない」と標記されます。注消統計がないため、閉店数は作れないと明示します。" } }, "whenToUse": "Relevant when working on: micro-business, 零售, finddata.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:market-flash-briefing", "slug": "skill-market-flash-briefing", "category": "skill", "marketCategory": null, "tags": [], "canonical": { "brief": "Condenses a piece of financial news/announcement/commentary into a one-page quick look: a three-line conclusion, five key points, and one open question, in a fixed output format." }, "human": { "zh-CN": { "name": "market-flash-briefing", "summary": "把一段财经新闻/公告/评论收敛成一页速览：三行结论、五个要点、一处存疑，输出固定版式。" }, "en": { "name": "market-flash-briefing", "summary": "Condenses a piece of financial news/announcement/commentary into a one-page quick look: a three-line conclusion, five key points, and one open question, in a fixed output format." }, "es": { "name": "market-flash-briefing", "summary": "Condensa una noticia/anuncio/comentario financiero en una vista rápida de una página: conclusión en tres líneas, cinco puntos clave y una duda abierta, con formato de salida fijo." }, "fr": { "name": "market-flash-briefing", "summary": "Condense une actualité/annonce/commentaire financier en un aperçu d'une page : conclusion en trois lignes, cinq points clés et un point laissé en doute, dans un format de sortie fixe." }, "ja": { "name": "market-flash-briefing", "summary": "金融ニュース/公告/評論を1ページのクイックビューに凝縮します：3行の結論、5つの要点、1つの疑問点。固定のレイアウトで出力します。" } }, "whenToUse": "把一段财经新闻/公告/评论收敛成一页速览：三行结论、五个要点、一处存疑，输出固定版式。", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:fd-vaas-publish-videos", "slug": "skill-fd-vaas-publish-videos", "category": "skill", "marketCategory": null, "tags": [ "wanxing", "vaas", "publish" ], "canonical": { "brief": "Publish videos to 6 platforms (Douyin/Xiaohongshu/Bilibili/Kuaishou/WeChat Channels/YouTube); vendored social-auto-upload + patchright; manual confirmation before publishing." }, "human": { "zh-CN": { "name": "fd-vaas-publish-videos", "summary": "把视频发布到 6 平台（抖音/小红书/B站/快手/视频号/YouTube），vendored social-auto-upload + patchright，发布前人工确认" }, "en": { "name": "fd-vaas-publish-videos", "summary": "Publish videos to 6 platforms (Douyin/Xiaohongshu/Bilibili/Kuaishou/WeChat Channels/YouTube); vendored social-auto-upload + patchright; manual confirmation before publishing." }, "es": { "name": "fd-vaas-publish-videos", "summary": "Publica vídeos en 6 plataformas (Douyin/Xiaohongshu/Bilibili/Kuaishou/WeChat Channels/YouTube); social-auto-upload + patchright incluidos (vendored); confirmación manual antes de publicar." }, "fr": { "name": "fd-vaas-publish-videos", "summary": "Publie des vidéos sur 6 plateformes (Douyin/Xiaohongshu/Bilibili/Kuaishou/Chaînes WeChat/YouTube) ; social-auto-upload + patchright embarqués (vendored) ; confirmation manuelle avant publication." }, "ja": { "name": "fd-vaas-publish-videos", "summary": "動画を6プラットフォーム（抖音/Douyin、小红书/Xiaohongshu、B站/Bilibili、快手/Kuaishou、视频号/WeChat Channels、YouTube）へ投稿します。同梱の social-auto-upload + patchright を使用し、投稿前に手動確認を行います。" } }, "whenToUse": "Relevant when working on: wanxing, vaas, publish.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:fd-vaas-pull-from-cloud", "slug": "skill-fd-vaas-pull-from-cloud", "category": "skill", "marketCategory": null, "tags": [ "wanxing" ], "canonical": { "brief": "Pull tenant-owned VAAS assets to local publish dirs" }, "human": { "zh-CN": { "name": "fd-vaas-pull-from-cloud", "summary": "Pull tenant-owned VAAS assets to local publish dirs" }, "en": { "name": "fd-vaas-pull-from-cloud", "summary": "Pull tenant-owned VAAS assets to local publish dirs" }, "es": { "name": "fd-vaas-pull-from-cloud", "summary": "Trae los activos VAAS propiedad del tenant a los directorios locales de publicación" }, "fr": { "name": "fd-vaas-pull-from-cloud", "summary": "Rapatrie les actifs VAAS appartenant au tenant vers les répertoires de publication locaux" }, "ja": { "name": "fd-vaas-pull-from-cloud", "summary": "テナント所有の VAAS アセットをローカルの公開ディレクトリに取り込みます" } }, "whenToUse": "Relevant when working on: wanxing.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "skill:fd-vaas-login", "slug": "skill-fd-vaas-login", "category": "skill", "marketCategory": null, "tags": [ "wanxing", "vaas" ], "canonical": { "brief": "VAAS login manager: scan-QR login for 6 video + 11 article platforms; login states stay on the user's machine" }, "human": { "zh-CN": { "name": "fd-vaas-login", "summary": "VAAS login manager: scan-QR login for 6 video + 11 article platforms; login states stay on the user's machine" }, "en": { "name": "fd-vaas-login", "summary": "VAAS login manager: scan-QR login for 6 video + 11 article platforms; login states stay on the user's machine" }, "es": { "name": "fd-vaas-login", "summary": "Gestor de inicio de sesión de VAAS: acceso por escaneo de código QR para 6 plataformas de vídeo + 11 de artículos; los estados de sesión permanecen en la máquina del usuario" }, "fr": { "name": "fd-vaas-login", "summary": "Gestionnaire de connexion VAAS : connexion par scan de QR pour 6 plateformes vidéo + 11 plateformes d'articles ; les états de connexion restent sur la machine de l'utilisateur" }, "ja": { "name": "fd-vaas-login", "summary": "VAAS ログインマネージャー：動画6 + 記事11プラットフォームのQRスキャンログイン。ログイン状態はユーザーのマシン上に保持されます" } }, "whenToUse": "Relevant when working on: wanxing, vaas.", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/skills-catalog", "updatedAt": "2026-11-02" } }, { "id": "mcp:airegistry-tools", "slug": "mcp-airegistry-tools", "category": "mcp", "marketCategory": null, "tags": [ "registry", "discovery", "search", "semantic-search", "tool-finder", "servers", "agents", "skills" ], "canonical": { "brief": "Provides tools to discover and list servers, agents, and skills in the AI Registry. Includes intelligent tool finder which uses semantic search to discover the most relevant tools across all registered services." }, "human": { "zh-CN": { "name": "AI Registry tools", "summary": "Provides tools to discover and list servers, agents, and skills in the AI Registry. Includes intelligent tool finder which uses semantic search to discover the most relevant tools across all registered services." }, "en": { "name": "AI Registry tools", "summary": "Provides tools to discover and list servers, agents, and skills in the AI Registry. Includes intelligent tool finder which uses semantic search to discover the most relevant tools across all registered services." }, "es": { "name": "AI Registry tools", "summary": "Provides tools to discover and list servers, agents, and skills in the AI Registry. Includes intelligent tool finder which uses semantic search to discover the most relevant tools across all registered services." }, "fr": { "name": "AI Registry tools", "summary": "Provides tools to discover and list servers, agents, and skills in the AI Registry. Includes intelligent tool finder which uses semantic search to discover the most relevant tools across all registered services." }, "ja": { "name": "AI Registry tools", "summary": "Provides tools to discover and list servers, agents, and skills in the AI Registry. Includes intelligent tool finder which uses semantic search to discover the most relevant tools across all registered services." } }, "whenToUse": "Provides tools to discover and list servers, agents, and skills in the AI Registry. Includes intelligent tool finder whi", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/mcp-catalog", "updatedAt": "2026-11-02" }, "tools": [ { "name": "list_services", "description": "List all MCP servers registered in the registry. Use search_registry\ninstead if you know what capability you need (faster, ranked results).\n\nEach server entry includes its endpoint URL, tools, and connection details.\nUse this for browsing the full catalog or when you need an unfiltered list." }, { "name": "list_agents", "description": "List all agents registered in the registry. Use search_registry\ninstead if you know what task you need an agent for (faster, ranked).\n\nAgents are autonomous services you can delegate tasks to. Each entry\nincludes the agent's URL, capabilities, and skills." }, { "name": "list_skills", "description": "List all skills registered in the registry. Use search_registry\ninstead if you know what workflow you need (faster, ranked results).\n\nSkills are reusable workflow instructions (like slash commands) that\nyou can load and execute. Use get_skill_content to retrieve the full\nmarkdown instructions for a discovered skill.\n\nReturns:\n Dictionary containing skills, total_count, and status" }, { "name": "get_skill_content", "description": "Retrieve the full instructions for a skill. Call this after finding a\nskill via search_registry to get its complete workflow markdown.\n\nThe returned content is a SKILL.md file containing step-by-step\ninstructions you can follow to execute the workflow. Some skills also\nhave companion resources (reference docs, scripts, configs) listed in\nthe manifest that you can fetch with the resource_path parameter." }, { "name": "search_registry", "description": "Discover AI assets (MCP servers, tools, agents, skills, and any custom\ncatalog types this registry defines) by describing what you need. Use this\nas your first step when you need a capability you don't currently have.\n\nResults include connection details so you can use the discovered assets:\n- Servers: have an endpoint_url field you can connect to directly as an\n MCP server (add it to your MCP client's server configuration)\n- Tools: individual capabilities within servers, with inputSchema\n- Agents: autonomous agents with a URL you can delegate tasks to\n- Skills: workflow instructions (use get_skill_content to fetch the full markdown)\n- Custom: records of a catalog type the registry administrator defined, each\n naming its type in entity_type. These are catalogued assets rather than\n endpoints, and there is no tool to fetch one: the search result below is\n the whole picture available here, carrying name, description, tags, owner\n and visibility. The path identifies the record for a caller that can reach\n the registry's HTTP API directly.\n\nWhen a useful MCP server is found, use the endpoint_url to add it to\nthe AI assistant's MCP configuration so its tools become available.\nFor Claude Code, run:\n claude mcp add --transport http --scope user \nNote: adding a server usually requires restarting the AI assistant\nsession for the new tools to take effect." }, { "name": "intelligent_tool_finder", "description": "DEPRECATED: Use search_registry instead. This tool will be removed in v1.26.0.\n\nSearch for tools using natural language semantic search." }, { "name": "healthcheck", "description": "Get registry health status and statistics.\n\nReturns:\n Dictionary containing health stats and status" } ], "toolCount": 7, "license": null, "provider": null, "endpoint": "https://mcp.finddatatech.cloud/airegistry-tools/mcp" }, { "id": "mcp:eco-mcp-fetch", "slug": "mcp-eco-mcp-fetch", "category": "mcp", "marketCategory": null, "tags": [ "ecosystem", "official" ], "canonical": { "brief": "[生态] modelcontextprotocol 官方参考实现（MIT）" }, "human": { "zh-CN": { "name": "网页抓取（托管版）", "summary": "托管在平台上的网页抓取：取回指定网址内容并转成可读文本。", "whenToUse": "不想在本机装进程、直接把网页内容拉回来时。" }, "en": { "name": "Web fetch (hosted)", "summary": "Hosted web fetch: retrieve a URL's content and convert it to readable text.", "whenToUse": "When you want page content without installing anything locally." }, "es": { "name": "Captura web (alojada)", "summary": "Captura web alojada: obtiene el contenido de una URL y lo convierte en texto legible.", "whenToUse": "Cuando quieres el contenido de una página sin instalar nada localmente." }, "fr": { "name": "Capture web (hébergée)", "summary": "Capture web hébergée : récupère le contenu d'une URL et le convertit en texte lisible.", "whenToUse": "Pour obtenir le contenu d'une page sans rien installer localement." }, "ja": { "name": "ウェブ取得（ホスト版）", "summary": "プラットフォーム上でホストされるウェブ取得：指定 URL の内容を取得し、読みやすいテキストに変換します。", "whenToUse": "ローカルに何も入れずにページ内容を取得したいとき。" } }, "whenToUse": "不想在本机装进程、直接把网页内容拉回来时。", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/mcp-catalog", "updatedAt": "2026-11-02" }, "tools": [ { "name": "fetch", "description": "Fetches a URL from the internet and optionally extracts its contents as markdown.\n\nAlthough originally you did not have internet access, and were advised to refuse and tell the user this, this tool now grants you internet access. Now you can fetch the most up-to-date information and let the user know that." } ], "toolCount": 1, "license": "MIT", "provider": "modelcontextprotocol", "endpoint": "https://mcp.finddatatech.cloud/eco-mcp-fetch/mcp" }, { "id": "mcp:eco-mcp-time", "slug": "mcp-eco-mcp-time", "category": "mcp", "marketCategory": null, "tags": [ "ecosystem", "official" ], "canonical": { "brief": "[生态] modelcontextprotocol 官方参考实现（MIT）" }, "human": { "zh-CN": { "name": "时间与时区（托管版）", "summary": "托管在平台上的时间服务：取当前时间、做时区换算。", "whenToUse": "不想在本机装进程、只需要算时间和时区时。" }, "en": { "name": "Time & timezone (hosted)", "summary": "Hosted time service: current time and timezone conversion.", "whenToUse": "When you only need time and timezone maths, with nothing installed locally." }, "es": { "name": "Hora y zona horaria (alojada)", "summary": "Servicio de hora alojado: hora actual y conversión de zonas horarias.", "whenToUse": "Cuando solo necesitas cálculos de hora y zona, sin instalar nada." }, "fr": { "name": "Heure et fuseau (hébergé)", "summary": "Service d'heure hébergé : heure courante et conversion de fuseaux horaires.", "whenToUse": "Pour des calculs d'heure et de fuseau uniquement, sans installation locale." }, "ja": { "name": "時刻とタイムゾーン（ホスト版）", "summary": "プラットフォーム上でホストされる時刻サービス：現在時刻の取得とタイムゾーン変換を行います。", "whenToUse": "ローカルに何も入れず、時刻とタイムゾーンの計算だけが必要なとき。" } }, "whenToUse": "不想在本机装进程、只需要算时间和时区时。", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/mcp-catalog", "updatedAt": "2026-11-02" }, "tools": [ { "name": "get_current_time", "description": "Get current time in a specific timezone" }, { "name": "convert_time", "description": "Convert time between timezones" } ], "toolCount": 2, "license": "MIT", "provider": "modelcontextprotocol", "endpoint": "https://mcp.finddatatech.cloud/eco-mcp-time/mcp" }, { "id": "mcp:facet-ops-mcp", "slug": "mcp-facet-ops-mcp", "category": "mcp", "marketCategory": null, "tags": [ "facet", "registry", "ops" ], "canonical": { "brief": "Facet ops MCP (registry face): agent catalog browse/register/update/toggle over the gateway AgentCard registry" }, "human": { "zh-CN": { "name": "谦面网关操作", "summary": "谦面网关的运营面：目录浏览、智能体名片、技能与服务的检索与检查。", "whenToUse": "在壹座或谦面里盘点供给目录、核对服务状态时。" }, "en": { "name": "Facet gateway operations", "summary": "Operations view of the Facet gateway: catalogue browsing, agent cards, and skill/service lookup and checks.", "whenToUse": "For auditing the supply catalogue or checking service status in Facet or the platform." }, "es": { "name": "Operaciones de la pasarela Facet", "summary": "Vista operativa de la pasarela Facet: navegación del catálogo, fichas de agentes y consulta de habilidades y servicios.", "whenToUse": "Para auditar el catálogo de suministro o comprobar el estado de un servicio." }, "fr": { "name": "Opérations de la passerelle Facet", "summary": "Vue d'exploitation de la passerelle Facet : navigation du catalogue, fiches d'agents, recherche et contrôle des compétences et services.", "whenToUse": "Pour auditer le catalogue d'offre ou vérifier l'état d'un service." }, "ja": { "name": "フェイスゲートウェイ操作", "summary": "フェイスの運用面：カタログ閲覧、エージェントカード、スキルとサービスの検索・確認を行います。", "whenToUse": "供給カタログの棚卸しやサービス状態の確認をするとき。" } }, "whenToUse": "在壹座或谦面里盘点供给目录、核对服务状态时。", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/mcp-catalog", "updatedAt": "2026-11-02" }, "tools": [ { "name": "facet_catalog_browse", "description": "浏览谦面市场 pack 目录（GET /api/agents）。\n\n以调用方用户身份执行——可见范围（公开/私有/分组）由 registry 按该用户判定。\nquery 支持名称/描述模糊匹配；分页 limit/offset（默认 20/0）。\n空结果如实返回空列表。" }, { "name": "facet_agent_read", "description": "读取网关 AgentCard 详情（GET /api/agents/{path}）。\n\npath 为网关条目路径（如注册名）；metadata_fields 可选，\n请求附加元数据字段（如 surface/marketplace）。404 如实报 not_found。" }, { "name": "facet_agent_publish", "description": "登记新 A2A AgentCard 进谦面网关目录（POST /api/agents/register）——写操作。\n\nregistration 按 registry 契约薄转发：必填 name / url / supportedProtocol；\n常用可选：description、is_proxied（true=走网关路由，需配 proxy_target_url）、\nproxy_target_url、proxy_streaming。需要 registry 侧 publish_agent 权限（403\n如实透传）。建议先 facet_agent_duplicate_check 查重。" }, { "name": "facet_agent_update", "description": "更新已有 AgentCard 定义（PUT /api/agents/{path}）——写操作。\n\nregistration 与 facet_agent_publish 同形（registry AgentRegistrationRequest）；\n仅条目属主/有权者可改（403 如实透传）。" }, { "name": "facet_agent_toggle", "description": "启用/停用网关条目（POST /api/agents/{path}/toggle）——显式写操作。\n\nenabled=True 启用（生成路由），False 停用（条目保留路由摘除）。仅属主/有权者。" }, { "name": "facet_agent_duplicate_check", "description": "登记前咨询式查重（POST /api/agents/check-duplicates）——恒 200，不阻塞。\n\n返回 collision_with（同 URL 精确命中）与 advisory_matches（相似命中）；\n有匹配也可继续登记（advisory 语义）。self_path 用于排除自身。" } ], "toolCount": 6, "license": null, "provider": null, "endpoint": "https://mcp.finddatatech.cloud/facet-ops-mcp/mcp" }, { "id": "mcp:fd-daas-mcp-cust1", "slug": "mcp-fd-daas-mcp-cust1", "category": "mcp", "marketCategory": null, "tags": [ "daas", "per-customer", "workspace" ], "canonical": { "brief": "FindData DAAS MCP — 第二客户工作区实例（per-customer isolation witness, facet-mcp-foundation-v1 3.6；独立 PVC，与共享实例 31861 数据互不可见）" }, "human": { "zh-CN": { "name": "fd-daas-mcp-cust1", "summary": "FindData DAAS MCP — 第二客户工作区实例（per-customer isolation witness, facet-mcp-foundation-v1 3.6；独立 PVC，与共享实例 31861 数据互不可见）" }, "en": { "name": "fd-daas-mcp-cust1", "summary": "FindData DAAS MCP — 第二客户工作区实例（per-customer isolation witness, facet-mcp-foundation-v1 3.6；独立 PVC，与共享实例 31861 数据互不可见）" }, "es": { "name": "fd-daas-mcp-cust1", "summary": "FindData DAAS MCP — 第二客户工作区实例（per-customer isolation witness, facet-mcp-foundation-v1 3.6；独立 PVC，与共享实例 31861 数据互不可见）" }, "fr": { "name": "fd-daas-mcp-cust1", "summary": "FindData DAAS MCP — 第二客户工作区实例（per-customer isolation witness, facet-mcp-foundation-v1 3.6；独立 PVC，与共享实例 31861 数据互不可见）" }, "ja": { "name": "fd-daas-mcp-cust1", "summary": "FindData DAAS MCP — 第二客户工作区实例（per-customer isolation witness, facet-mcp-foundation-v1 3.6；独立 PVC，与共享实例 31861 数据互不可见）" } }, "whenToUse": "FindData DAAS MCP — 第二客户工作区实例（per-customer isolation witness, facet-mcp-foundation-v1 3.6；独立 PVC，与共享实例 31861 数据互不可见）", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/mcp-catalog", "updatedAt": "2026-11-02" }, "tools": [ { "name": "alerts_list_series", "description": "Distinct (source, function_name, indicator) from observations + scraw_* tables,\neach with a row count + latest date. Alert-scoped discovery (not a SQL browser)." }, { "name": "alerts_get_series_latest", "description": "Return the latest `limit` rows of a series ordered by date_column desc." }, { "name": "alerts_create_alert_rule", "description": "Create a trigger rule over a DB series.\n\n`condition` is a safe DSL (no eval): `latest > 70`, `crosses_above(70)`,\n`pct_change(5) > 0.05`, `latest > 70 and prev <= 70`, etc. `channels` is a\nlist of names ([\"telegram\",\"slack\"]) or a {name: override} object. All\nchannel credentials live in root .env under ALERTS_*." }, { "name": "alerts_list_alert_rules", "description": "Return all alert rules." }, { "name": "alerts_get_alert_rule", "description": "Return one rule by name." }, { "name": "alerts_update_alert_rule", "description": "Update a rule's fields. Only provided fields change. The name cannot be renamed." }, { "name": "alerts_delete_alert_rule", "description": "Delete a rule. Its alert_events rows are removed via FK CASCADE." }, { "name": "alerts_list_channels", "description": "Return {name, configured, missing_keys} for each of the 7 channels\n(telegram, discord, slack, twitter, dingtalk, feishu, wecowork).\n\nNever returns credential values — only the names of missing keys." }, { "name": "alerts_run_alert_rule", "description": "Ad-hoc: evaluate one rule now and dispatch if it fires.\n\nRespects fire_mode + cooldown like the cron path. Use to test a rule on demand." }, { "name": "alerts_list_events", "description": "Recent alert_events (optionally for one rule), newest first." }, { "name": "cron_create_schedule", "description": "Create a new schedule with a cron expression and task." }, { "name": "cron_list_schedules", "description": "List all schedules." }, { "name": "cron_get_schedule", "description": "Get details for a single schedule." }, { "name": "cron_delete_schedule", "description": "Delete a schedule permanently." }, { "name": "cron_pause_schedule", "description": "Pause a schedule (disable without deleting)." }, { "name": "cron_resume_schedule", "description": "Resume a paused schedule." }, { "name": "cron_run_now", "description": "Execute a scheduled task immediately (one-shot)." }, { "name": "cron_list_executions", "description": "List execution history, optionally filtered by schedule_id." }, { "name": "cron_list_tasks", "description": "List available task types in the registry." }, { "name": "cron_create_task", "description": "Register a new task in the database. The command is a shell command or script path.\n\nExample: create_task(name=\"fetch_mofcom\", command=\"python3 scripts/fetch_mofcom_news.py --json\")" }, { "name": "cron_delete_task", "description": "Delete a task from the database by name." }, { "name": "cron_list_db_tasks", "description": "List all tasks stored in the database (user-defined, no code changes needed)." }, { "name": "cron_update_task", "description": "Update an existing DB task's command, description, or timeout. Only provided fields are updated." }, { "name": "composite_list", "description": "List all defined composites (name + description)." }, { "name": "composite_create", "description": "Create a new composite. Raises if the name already exists." }, { "name": "composite_list_upstreams", "description": "List the upstreams defined for a composite." }, { "name": "composite_add_upstream", "description": "Add an upstream MCP server to a composite.\n\nFor stdio: provide command, args, cwd (env optional).\nFor http: provide url." }, { "name": "composite_remove_upstream", "description": "Remove an upstream (and its selected tools) from a composite." }, { "name": "composite_list_available_tools", "description": "List the tools an upstream currently exposes (live), optionally filtered." }, { "name": "composite_add_tool", "description": "Select an upstream tool to be exposed (proxied) by the composite.\n\nThe tool is served as _ on next start.\n(alias is accepted but unused in v1.)" }, { "name": "composite_remove_tool", "description": "Remove a selected tool from a composite." }, { "name": "composite_list_tools", "description": "List the tools currently selected for a composite." }, { "name": "composite_add_chained_tool", "description": "Define a chained tool: a linear pipeline of upstream tool calls.\n\nEach step is {\"upstream\": , \"tool\": , \"input\": {param: value}}.\nInput values may be literals or references: \"$prev.\" (prior step's\nresult) or \"$step[N].\" (any prior step's result). Linear only — no\nbranching/conditionals." }, { "name": "composite_remove_chained_tool", "description": "Remove a chained tool from a composite." }, { "name": "composite_list_chained_tools", "description": "List the chained tools defined for a composite." }, { "name": "composite_create_manifest", "description": "Create a composite manifest: composite + upstreams + tools + workflows + prompt in one call." }, { "name": "composite_update_manifest", "description": "Partially update a composite manifest. Only provided fields change.\n\nupstreams/tools (when given) REPLACE the existing sets wholesale." }, { "name": "composite_delete_manifest", "description": "Delete a composite manifest (cascades upstreams/tools/chains)." }, { "name": "composite_list_manifests", "description": "List all composites as manifests (name, upstreams, tools, workflows, prompt)." }, { "name": "daas_list_sources", "description": "List all configured DAAS data sources with function counts.\n\nReturns sources like akshare, worldbank, ckan, cnstats\nwith install status and available function counts." }, { "name": "daas_search_functions", "description": "Search DAAS functions by name, category, or description across all sources." }, { "name": "daas_get_function_detail", "description": "Get full details for a DAAS function — parameters, output columns, description." }, { "name": "daas_list_categories", "description": "List all DAAS function categories with counts, optionally filtered by source." }, { "name": "daas_fetch_data", "description": "Execute a DAAS data function and return results as JSON.\n\nRoutes to the correct source adapter based on function name prefix." }, { "name": "daas_create_datasource", "description": "Create a new managed datasource." }, { "name": "daas_update_datasource", "description": "Update mutable fields of a datasource, identified by name or id.\n\nPass clear_category=true to unset the category (move to root level).\nPass score= to set the default score; pass clear_score=true to reset\nthe score back to NULL (unset). Only supplied fields are changed." }, { "name": "daas_delete_datasource", "description": "Delete a datasource and cascade-delete its forms, sections, and\ncollection-item references. Identify by name or id." }, { "name": "daas_create_category", "description": "Create a category, optionally under a parent (building the tree)." }, { "name": "daas_move_category", "description": "Re-parent a category. Rejects moves that would create a cycle\n(into own descendant) or self-parenting. Pass parent_id=null to make\nit a root." }, { "name": "daas_delete_category", "description": "Delete a category. Rejected if it has child categories. Datasources\nassigned to it are orphaned to root level (category_id set to null)." }, { "name": "daas_get_category_tree", "description": "Return the category tree (or a subtree from root_id) as a nested\nstructure, each node annotated with datasource_count." }, { "name": "daas_add_form", "description": "Add a form (e.g. '10-K', '8-K') to a datasource." }, { "name": "daas_add_section", "description": "Add a section to a form, carrying an extraction instruction." }, { "name": "daas_list_forms", "description": "List all forms of a datasource, each with nested sections\n(including instruction)." }, { "name": "daas_create_collection", "description": "Create a named datasource collection." }, { "name": "daas_add_to_collection", "description": "Add a datasource (or a specific datasource-section) to a collection.\n\nOmit section_name to add the whole datasource; supply section_name to\nadd only that section. Optional score sets a per-collection score override\non the new item (NULL = inherit the datasource's default score)." }, { "name": "daas_list_collection_items", "description": "List a collection's items, each resolved to source name and (if set)\nsection name + instruction." }, { "name": "daas_remove_from_collection", "description": "Remove an item from a collection by (collection, source, optional\nsection). Omit section_name to remove the whole-datasource item." }, { "name": "daas_list_collections", "description": "List every datasource collection with its item count." }, { "name": "daas_rename_collection", "description": "Rename a collection. The new name must be unique. Items are preserved\n(they reference by collection_id)." }, { "name": "daas_update_collection", "description": "Partially update a collection's name and/or description.\n\nAt least one of `new_name` / `description` must be provided; omitted\nfields are left unchanged. When `new_name` differs from the current\nname it must be unique. Raises if the collection is not found." }, { "name": "daas_delete_collection", "description": "Delete a collection. Cascades to its `datasource_collection_items`\nrows (datasources themselves are untouched)." }, { "name": "daas_reorder_collection_items", "description": "Rewrite the sort_order of items in `collection_name` to match the\ngiven list. `ordered_item_ids` MUST include every existing item id in\nthe collection exactly once — partial reorders are rejected." }, { "name": "daas_set_collection_item_score", "description": "Set or clear the per-collection score override on an existing collection\nitem, identified by (collection_name, source_name, optional section_name).\n\nPass score= to set the override; pass score=null (omit) to clear it\nso the item falls back to the datasource's default score. Returns the\nupdated item with its resolved effective score, the raw item_score override,\nand the source_default_score." }, { "name": "daas_search_datasources", "description": "Search/filter datasources across levels — category (with optional\nsubtree), source, form, and section — every filter optional, plus a\nfree-text query across source label/description, form label, and\nsection name/instruction." }, { "name": "daas_search_entities", "description": "Search entities (stocks + countries) by name, ticker, code, or alias." }, { "name": "daas_get_entity", "description": "Get full detail for one entity, including aliases and metadata.\n\nPer D5/3.5 the entity master is fd-open-data-mcp. Resolve by natural\nkey (entity_type, code) when given (gateway-first, local fallback);\notherwise fall back to entity_id (local)." }, { "name": "daas_list_entities", "description": "List entities filtered by type / exchange / country, paginated." }, { "name": "daas_get_entity_coverage", "description": "For each datasource linked to the entity, return the identifier to\nuse, the available sections (routing instructions = how to get the\ndata, with an identifier-prefilled variant), and the column count/list\naggregated from daas_function_columns. Sources without registered\nfunctions get a column_hint naming the sibling MCP + tool.\n\nPer D5/3.5 entity resolution proxies through fd-open-data-mcp. Accept\na natural key (entity_type, code) as an alternative to entity_id.\n\nThis answers: \"I have company X — which datasources cover it, how many\ncolumns can I get, and how do I fetch it?\"" }, { "name": "daas_link_entity_datasource", "description": "Link an entity to a daas datasource, recording the identifier to use\ninside that datasource (e.g. for AAPL → yfinance: 'AAPL'). Upserts on\n(entity_id, source_name)." }, { "name": "daas_unlink_entity_datasource", "description": "Delete the link between an entity and a datasource." }, { "name": "daas_create_entity_collection", "description": "Create a named entity collection (watchlist / portfolio)." }, { "name": "daas_list_entity_collections", "description": "List every entity collection with its item count." }, { "name": "daas_get_entity_collection", "description": "Get one collection with its current member entities (ordered)." }, { "name": "daas_update_entity_collection", "description": "Partially update an entity collection's name and/or description and/or\nrule. At least one field must be provided. `clear_rule=True` resets the\nrule to NULL (manual collection)." }, { "name": "daas_delete_entity_collection", "description": "Delete a collection and cascade to its items + changes." }, { "name": "daas_add_entity_to_collection", "description": "Add an entity to a collection. Resolves the entity by `entity_id`, or\nby `(entity_type, code)`. Records an `add_in` event in\n`entity_collection_changes` (source='manual'). No-op (no event recorded)\nif the entity is already a member." }, { "name": "daas_remove_entity_from_collection", "description": "Remove an entity from a collection. Records a `remove_out` event in\n`entity_collection_changes` (source='manual'). No-op (no event recorded)\nif the entity is not a member." }, { "name": "daas_list_entity_collection_items", "description": "List the current members of a collection, ordered by sort_order, each\nenriched with full entity detail (code, name, ticker, exchange, ...)." }, { "name": "daas_reorder_entity_collection_items", "description": "Rewrite member sort_order to match the given ordered list of\n`entity_collection_items.id`. Must contain exactly the current item ids." }, { "name": "daas_list_entity_collection_changes", "description": "Query the add-in / remove-out audit log (`entity_collection_changes`),\nnewest first, each enriched with collection name + entity code/name." }, { "name": "daas_sync_entity_collection", "description": "Re-derive the member set for a rule-based collection by applying its\n`rule_json` to the `entities` table, diff vs current members, apply\nadd_in for new matches and remove_out for non-matches (source='cron'),\nand record every transition. Returns `{added, removed, unchanged}`.\nA manual collection (rule_json=NULL) is a no-op." }, { "name": "daas_create_indicator_collection", "description": "Create a named indicator collection (a reusable bundle of indicators)." }, { "name": "daas_list_indicator_collections", "description": "List every indicator collection with its item count." }, { "name": "daas_get_indicator_collection", "description": "Get one collection with its current items (ordered, each with the\nresolved effective score + raw item/indicator/datasource scores)." }, { "name": "daas_update_indicator_collection", "description": "Partially update an indicator collection's name, description, and/or\nrule. At least one field must be provided. `clear_rule=True` resets\n`rule_id` to NULL (manual collection)." }, { "name": "daas_delete_indicator_collection", "description": "Delete a collection and cascade to its items + changes." }, { "name": "daas_add_indicator_to_collection", "description": "Add an indicator to a collection. Resolves the indicator by `name`." }, { "name": "daas_remove_indicator_from_collection", "description": "Remove an indicator from a collection. Records a `remove_out` event in\nthe audit log. No-op (`not_member`) if the indicator is not in the collection." }, { "name": "daas_list_indicator_collection_items", "description": "List the items of a collection, ordered by sort_order, each with the\nresolved effective score + raw item/indicator/datasource scores." }, { "name": "daas_reorder_indicator_collection_items", "description": "Rewrite item sort_order to match the given ordered list of\n`indicator_collection_items.id`. Must contain exactly the current item\nids of this collection — partial reorders are rejected." }, { "name": "daas_set_indicator_collection_item_score", "description": "Set or clear the per-collection score override on an existing item,\nidentified by (collection_name, indicator_name).\n\n`score` = a float sets the override; `score = null` clears it (the item\nthen inherits the indicator's default `indicator_rules.score`, which\nitself inherits the datasource default when NULL). Returns the updated\nitem dict with the resolved effective score." }, { "name": "daas_list_indicator_collection_changes", "description": "Query the add-in / remove-out audit log (`indicator_collection_changes`),\nnewest first. Each row is enriched with the collection name." }, { "name": "daas_sync_indicator_collection", "description": "Re-derive the member set for a rule-based indicator collection by\nevaluating its rule (target='indicator_names') via the RuleEngine, diff vs\ncurrent members, apply add_in/remove_out (source='cron'), and record every\ntransition. Returns `{added, removed, unchanged}`. A manual collection\n(rule_id NULL) is a no-op." }, { "name": "daas_create_pipeline_collection", "description": "Create a named pipeline collection." }, { "name": "daas_list_pipeline_collections", "description": "Every pipeline collection with its item count." }, { "name": "daas_get_pipeline_collection", "description": "One pipeline collection with all its items." }, { "name": "daas_delete_pipeline_collection", "description": "Unwire cron for every item, then delete the collection (FK CASCADE removes items)." }, { "name": "daas_list_pipeline_items", "description": "List items, optionally filtered to one collection." }, { "name": "daas_add_pipeline_item", "description": "Validate + store an item. If enabled and backfill, run history backfill\nthen register the cron-mcp schedule. Returns {item, backfill, cron}." }, { "name": "daas_remove_pipeline_item", "description": "Unwire cron then delete the item row. scraw_ is left intact." }, { "name": "daas_enable_pipeline_item", "description": "Backfill + register cron for a disabled item; set enabled=1." }, { "name": "daas_disable_pipeline_item", "description": "Delete the cron schedule (keep the task row); set enabled=0." }, { "name": "daas_update_pipeline_item", "description": "Update mutable fields. cron/timezone change re-syncs the schedule.\nenabled toggle runs enable/disable. arguments_json change does NOT backfill." }, { "name": "daas_sync_pipeline_cron", "description": "Re-apply cron wiring for all enabled items; remove schedules for disabled ones." }, { "name": "daas_list_models", "description": "Return configured models with a `vision` flag each. API keys are never returned." }, { "name": "daas_list_source_tables", "description": "Return scraped source-data tables (name LIKE 'scraw_%') with row counts + columns." }, { "name": "daas_create_rule", "description": "Create a rule." }, { "name": "daas_list_rules", "description": "List all rules, optionally filtered by rule_type." }, { "name": "daas_get_rule", "description": "Return one rule by name." }, { "name": "daas_update_rule", "description": "Partially update a rule. At least one field must be provided. The rule\nname cannot be renamed." }, { "name": "daas_delete_rule", "description": "Delete a rule. Referencing collections' rule_id is set to NULL; the\nrule's process_results rows are removed." }, { "name": "daas_run_rule", "description": "Evaluate + persist. For `target='rows'` (llm): incremental extraction\ninto process_results + advance last_rowid. For member targets: returns the\nderived set (membership is applied by daas_sync_*_collection, not here)." }, { "name": "daas_test_rule", "description": "Dry-run: evaluate the rule WITHOUT persisting. For member targets\n(entity_ids/indicator_names) returns the derived set. For `target='rows'`\n(llm), extracts from a single source row (no process_results write) as a\nsample." }, { "name": "daas_extract_text", "description": "Extract structured records from (possibly long) text.\n\nLong text is chunked (no truncation): per-chunk extraction then a merge pass." }, { "name": "daas_extract_image", "description": "Extract structured records from an image via a vision-capable model.\n\n`image` accepts a local file path, an http(s):// URL, or raw base64." }, { "name": "daas_extract_file", "description": "Read a local .txt/.md/.pdf and extract structured records." }, { "name": "daas_list_indicator_ops", "description": "Return the fixed math-op catalog with each op's required params." }, { "name": "daas_create_indicator", "description": "Create an indicator rule bound to a datasource + source table." }, { "name": "daas_list_indicators", "description": "Return all indicator rules (each with `score` + `effective_default_score`)." }, { "name": "daas_get_indicator", "description": "Return one indicator rule by name (with `score` + `effective_default_score`)." }, { "name": "daas_update_indicator", "description": "Update an indicator rule's fields. Only provided fields change. The\nrule name cannot be renamed. Pass `params` to replace the op params.\n\n`score`: a float sets the indicator's default score. `clear_score=True`\nclears it back to NULL (inherit the datasource's `sources.score`).\nMirrors `update_datasource(score=…, clear_score=True)`." }, { "name": "daas_set_indicator_score", "description": "Set the indicator's default `score` (float), or clear it (None → inherit\nthe datasource's `sources.score`). Returns the updated indicator dict with\n`effective_default_score`." }, { "name": "daas_delete_indicator", "description": "Delete an indicator rule. observations rows it produced are NOT removed\n(soft reference) — they remain identifiable via their `metadata.rule_name`." }, { "name": "daas_run_indicator", "description": "Full-recompute the indicator over its source table → observations.\n\nReads the whole source table, computes the op, and upserts every\n(date, value) into the daas `observations` table keyed on\n(source=datasource, function_name, indicator=indicator_name, date).\nIdempotent on re-run via the observations unique constraint." }, { "name": "daas_calculate_indicator", "description": "Ad-hoc: compute an indicator over a source table without persisting.\n\nValidates table/columns/op/params, reads the series, computes, and returns\n{indicator, dates, values, count}. Writes nothing (no rule, no observations)." }, { "name": "dashboard_list_databases", "description": "List all known SQLite databases in the MCP directory with table counts." }, { "name": "dashboard_query_table", "description": "Run a read-only query against a database table. Returns rows as JSON." }, { "name": "dashboard_list_datasources", "description": "List managed datasources from daas.db (schema managed by mcp/models/models.py)." }, { "name": "dashboard_get_executions", "description": "Get cron execution history from a database." }, { "name": "dashboard_get_stats", "description": "Get aggregated statistics across all MCP databases." }, { "name": "dashboard_register", "description": "Register or update (upsert by slug) a standalone HTML dashboard in the\n`dashboards` table. After writing the row, regenerates index.html + daas.md\nfrom the DB. `source_tables`/`entity_coverage`/`time_range`/`chart_config`\naccept a JSON string or a list/dict. Returns the stored row (with `action`\n= inserted|updated) or `{\"error\": ...}`." }, { "name": "dashboard_list", "description": "List every registered dashboard (name + slug + intro + file_url), oldest\nfirst. Returns a JSON array." }, { "name": "dashboard_get", "description": "Get one dashboard's full metadata by slug: name, intro, source_tables,\nentity_coverage, time_range, refresh_cadence, chart_config, file_path,\nfile_url. Returns `{\"error\": \"dashboard '' not found\"}` if absent." }, { "name": "dashboard_search", "description": "Case-insensitive keyword search over each dashboard's name + intro +\nrefresh_cadence + source_tables. Returns matching `{slug, name, intro}`\nentries, or an empty array if none match." }, { "name": "dashboard_update", "description": "Patch one or more fields of an existing dashboard (by slug). Only the\nfields you pass are changed; the rest are left as-is. Regenerates index.html\n+ daas.md after the update. Returns the updated row or `{\"error\": ...}`." }, { "name": "dashboard_delete", "description": "Delete a dashboard row by slug and regenerate index.html + daas.md. The\nHTML file itself is NOT removed (only the registry row). Returns\n`{\"deleted\": slug}` or `{\"error\": ...}`." }, { "name": "gateway_list_data_mcps", "description": "List the data-fetch MCP upstreams gateway-mcp can route to.\n\nEach entry is a row from the `gateway_upstreams` table (name, transport,\nenabled, description). Disabled upstreams are hidden unless\n`include_disabled=True`." }, { "name": "gateway_list_data_mcp_tools", "description": "List the tools exposed by a data-fetch MCP upstream (live).\n\nReuses the persistent client from the connection pool — the client stays\nalive across calls and is only recreated when the upstream config changes.\nReturns each tool's name + description (+ parameters when available)." }, { "name": "gateway_call_data_mcp", "description": "Call a tool on a data-fetch MCP upstream and return its result.\n\nConnects to the named upstream via the persistent client pool, invokes\nthe named tool with the JSON-deserialized `arguments`, and returns the\nupstream's result. The client stays alive across calls and is only\nrecreated when the upstream config changes." }, { "name": "gateway_add_data_mcp", "description": "Add or update a data-fetch MCP upstream (idempotent upsert by name)." }, { "name": "gateway_remove_data_mcp", "description": "Remove a data-fetch MCP upstream from the registry." }, { "name": "gateway_get_data_mcp", "description": "Get one data-fetch MCP upstream's full launch config." }, { "name": "gateway_gateway_health", "description": "Health-probe each enabled gateway upstream and auto-flip transport.\n\nFor each enabled upstream in ``gateway_upstreams``:\n\n- **http** transport: ping the row's ``url``. On failure AND the row\n carries a stdio ``command``, flip ``transport`` to ``stdio`` (degraded\n mode — mirrors the client-pool fallback so the next ``get_client()``\n launches the subprocess). On success, no change.\n- **stdio** transport with a ``url`` set (i.e. it was flipped from http by\n a prior fallback): ping the ``url``. On success flip ``transport`` back\n to ``http`` (recovered). On failure, leave as stdio.\n- **stdio**-only (no ``url``): skipped (nothing to probe).\n\nReturns a per-upstream status report. Never raises — network errors are\ncaptured as degraded states so the probe is safe to call from selfcheck." }, { "name": "research_create", "description": "Create a persisted research bundle. Named collections are attached by\nname; with ``create_missing=True`` missing entity/indicator/pipeline\ncollections are created empty. A ``dashboard_slug`` must already exist\n(the tool does not build dashboard HTML). ``component_refs`` is a JSON\nobject string: ``{\"rules\": [...], \"scraw_tables\": [...], \"indicators\": [...]}``." }, { "name": "research_get", "description": "Return a research + resolved bundle (live counts, dashboard url,\npipeline status, dangling-component warnings)." }, { "name": "research_list", "description": "List every research (optionally filtered by status) with attached\ncomponent names." }, { "name": "research_update", "description": "Partially update a research. ``None`` means leave a field unchanged;\nto detach a named component use ``remove_component``. Any newly-set\ncomponent name is validated to exist. ``component_refs`` (JSON string)\nreplaces the auxiliary refs." }, { "name": "research_delete", "description": "Delete a research. Removes the on-disk report file. With\n``remove_pipeline=True`` (default) deletes the owned pipeline collection\n+ orphaned cron schedules/tasks. With ``remove_dashboard=True`` (default\nFalse) deletes the referenced dashboard via the sibling tool. Shared\nentity/indicator collections, rules, scraw tables, and observations are\nNEVER deleted." }, { "name": "research_generate_report", "description": "Assemble a markdown report from the bundle's live components, persist it\ninto ``report_md`` and write it to ``researches/.md`` (dir from\n``RESEARCH_DIR``, default repo-root ``researches/``). Regeneration\noverwrites both. Returns ``{name, report_path, char_count}``." }, { "name": "research_refresh", "description": "Re-run the research's data pipeline: recompute each indicator in the\nindicator collection (via ``daas_run_indicator``), sync rule-based\ncollections (via ``daas_sync_entity_collection``/``daas_sync_indicator_collection``),\nand report each pipeline item's current status. Returns a per-component\nstatus report; per-item errors are reported, not raised." }, { "name": "research_add_component", "description": "Attach an existing component to a research by name. ``component_type`` ∈\nentity_collection | indicator_collection | dashboard | pipeline | rule |\nscraw_table | indicator. The underlying object is not created - it must\nalready exist (for the named kinds)." }, { "name": "research_remove_component", "description": "Detach a component from a research. The underlying object is NOT\ndeleted. For named kinds, clears the research column; for auxiliary refs,\nremoves the name from ``component_refs``." }, { "name": "workflow_register", "description": "Register a new workflow manifest (YAML). Version bumps on name collision\nunless the manifest declares its own ``version``." }, { "name": "workflow_get", "description": "Get a workflow by name (latest version by default)." }, { "name": "workflow_list", "description": "List every workflow (name, version, description, enabled, step_count)." }, { "name": "workflow_update", "description": "Patch a workflow. Only provided fields change. ``version`` defaults to\nthe latest version of ``name``." }, { "name": "workflow_delete", "description": "Delete a workflow. Omit ``version`` to delete every version of ``name``.\nSteps/runs cascade via FK." }, { "name": "workflow_run", "description": "Run a workflow manifest. ``params_json`` merges over the manifest's\ndeclared ``params``. Returns run id + status; a paused run carries a\n``resume_token`` for ``resume``." }, { "name": "workflow_inspect", "description": "Validate a stored manifest and return its parsed step plan (no execution)." }, { "name": "workflow_resume", "description": "Resume a paused run. ``approved=False`` marks it failed." } ], "toolCount": 161, "license": null, "provider": null, "endpoint": "https://mcp.finddatatech.cloud/fd-daas-mcp-cust1/mcp" }, { "id": "mcp:fd-health-config", "slug": "mcp-fd-health-config", "category": "mcp", "marketCategory": null, "tags": [], "canonical": { "brief": "finddata 中央库健康巡检配置(总闸/限流)只读 MCP — fd_open_data.public.health_config" }, "human": { "zh-CN": { "name": "健康巡检配置", "summary": "读取中央库健康巡检的总闸与限流配置，只读。", "whenToUse": "核对巡检是否开启、限流阈值是多少时。" }, "en": { "name": "Health config", "summary": "Read the central health-check master switch and rate-limit configuration; read-only.", "whenToUse": "To confirm whether checks are enabled and what the rate limits are." }, "es": { "name": "Configuración de salud", "summary": "Lectura del interruptor principal de comprobaciones y de los límites de tasa; solo lectura.", "whenToUse": "Para confirmar si las comprobaciones están activas y cuáles son los límites." }, "fr": { "name": "Configuration de santé", "summary": "Lecture de l'interrupteur principal des contrôles de santé et des limites de débit ; lecture seule.", "whenToUse": "Pour vérifier si les contrôles sont actifs et quelles sont les limites." }, "ja": { "name": "ヘルス設定", "summary": "中央のヘルスチェックの主スイッチとレート制限設定を読み取ります。読み取り専用です。", "whenToUse": "チェックの有効性やレート制限の値を確認するとき。" } }, "whenToUse": "核对巡检是否开启、限流阈值是多少时。", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/mcp-catalog", "updatedAt": "2026-11-02" }, "tools": [ { "name": "health_config_get", "description": "读取 finddata 中央库健康巡检配置（fd_open_data.public.health_config 全量键值）：master_switch（总闸，false=纯巡检不出修复单）、max_daily_tickets（每日工单上限）、lookback_hours、thresholds、excluded_sources、expected_period_overrides；updated_at 为最近变更时间。" } ], "toolCount": 1, "license": null, "provider": null, "endpoint": "https://mcp.finddatatech.cloud/fd-health-config/mcp" }, { "id": "mcp:fd-legal-search-mcp", "slug": "mcp-fd-legal-search-mcp", "category": "mcp", "marketCategory": null, "tags": [ "legal", "regulations", "chinese-law", "finddata", "lex" ], "canonical": { "brief": "识律法律检索 MCP（lex-platform-v1）：法条检索与条级引用解析（law_db 全目录 + Qdrant law_chunks_v2 语义检索）、类案检索（wenshu_db 裁判文书，RustFS zstd 块正文）、指导案例检索、五源法律语料——每个结果携带 provenance 三件套（source/stable_id/as_of），未命中如实返回。识律平台 lex.finddatatech.cloud 对外数据入口，行/次双单位计量。" }, "human": { "zh-CN": { "name": "法律检索", "summary": "法规、案例与指导案例的检索与读取，附文书库检索入口。", "whenToUse": "找法条、找类案或查裁判观点时。" }, "en": { "name": "Legal search", "summary": "Search and read statutes, cases, and guiding cases, with an entry to the judgment corpus.", "whenToUse": "When locating a statute, similar cases, or a judicial view." }, "es": { "name": "Búsqueda jurídica", "summary": "Búsqueda y lectura de normativa, casos y casos guía, con acceso al corpus de sentencias.", "whenToUse": "Para localizar una norma, casos similares o un criterio judicial." }, "fr": { "name": "Recherche juridique", "summary": "Recherche et lecture de textes légaux, d'affaires et d'affaires de référence, avec accès au corpus de jugements.", "whenToUse": "Pour trouver un texte, des affaires similaires ou une position judiciaire." }, "ja": { "name": "法令検索", "summary": "法令・判例・指導性案例の検索と読解に加え、判決文コーパスへの入口を提供します。", "whenToUse": "条文、類似判例、裁判所の判断を探すとき。" } }, "whenToUse": "找法条、找类案或查裁判观点时。", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/mcp-catalog", "updatedAt": "2026-11-02" }, "tools": [ { "name": "law_search", "description": "Search FindData law titles, ranked exact > prefix > contains.\n\n法规库全目录检索（law_db.laws，含别名治理）；category 接受同义拼写。\n结果逐条携带 provenance（source=law_db.laws, stable_id=law_id）。\n与 business-mcp law_search 等价迁移（lex-platform-v1 task 7.1）。" }, { "name": "law_read", "description": "Read one law's metadata and text chunks by law_id.\n\n返回法规详情与条级 chunk 结构，供正文阅读与条级引用使用。\n与 business-mcp law_read 等价迁移（lex-platform-v1 task 7.1）。\n\n条文按窗口返回：单次最多 ``limit`` 条（默认 200，上限 500），从\n``offset`` 起算。整部法规一次全量回传会让大法典独占调用方上下文；\n``total_chunks`` 与 ``next_offset`` 给出续读位置，分段拼接即全文。" }, { "name": "law_semantic_search", "description": "法条语义检索：Qwen3-Embedding-0.6B 查询向量 -> Qdrant law_chunks_v2。\n\n按语义相似度返回条级 chunk，逐条携带 provenance\n（source=law_db.law_chunks_v2, stable_id=point id, as_of=导出日期）。\n实现跟踪：lex-platform-v1 task 1.2。" }, { "name": "law_reference", "description": "条级引用解析：把「《民法典》第一百四十三条」类引用解析到法规目录条级定位。\n\n无法解析时如实返回 not_found（含可读原因），绝不猜测。\n实现跟踪：lex-platform-v1 task 1.2。" }, { "name": "wenshu_search", "description": "Search judgment titles (3,700万篇裁判文书), ranked exact > prefix > contains.\n\n案号内嵌于标题（如「（2024）京01民初1号」），无需单独案号参数。\n结果逐条携带 provenance（source=wenshu_db.judgments, stable_id=doc_id）。\n与 business-mcp wenshu_search 等价迁移（lex-platform-v1 task 7.1）。" }, { "name": "wenshu_read", "description": "Read one judgment's full text (zstd blocks on RustFS).\n\n正文按块解压读取；条目存在而正文缺失时如实标注，绝不编造正文。\n与 business-mcp wenshu_read 等价迁移（lex-platform-v1 task 7.1）。\n\n长正文按窗口返回：单次最多 ``max_chars`` 字符（默认 12000，上限 40000）。\n超出时 ``content_truncated=true`` 且给 ``next_offset``——把它作为\n``offset`` 再调一次即可续读，分段按字符切分，拼接后与原文逐字符一致。" }, { "name": "guide_cases_search", "description": "Search guiding-case titles, ranked exact > prefix > contains.\n\ndoc_type / reference_status 为封闭词表过滤（非法值如实拒绝）。\n结果逐条携带 provenance（source=law_db.guide_cases, stable_id=case_id）。\n与 business-mcp guide_cases_search 等价迁移（lex-platform-v1 task 7.1）。" }, { "name": "guide_cases_read", "description": "Read one guiding case's full structured sections by case_id.\n\n与 business-mcp guide_cases_read 等价迁移（lex-platform-v1 task 7.1）。" }, { "name": "legal_docs_search", "description": "Search legal-document titles across the five corpora (47,256 篇).\n\nRanked exact > prefix > contains, case-insensitive. `source` is a closed\nvocabulary of the five corpus codes — an unknown value is rejected with\n`invalid_filter` plus the valid list, never silently matched to nothing.\n`doc_type` is an equality filter. At most 200 rows.\n\nDocument bodies are never returned here; `has_content` says whether\n`legal_docs_read` can serve one. Every hit carries provenance\n(source=law_db.legal_documents, stable_id=doc_id, as_of=入库时间)." }, { "name": "legal_docs_read", "description": "Read one legal document by `doc_id`, body included only when fetched.\n\nReturns full metadata (including `source_url` and the crawler `metadata`\njsonb minus its internal `object_keys`) plus `content_available`. When\nthe body never made it into the corpus the answer carries an honest\n`note` and the `content` key is **absent** — not an empty string.\nUnknown ids return `not_found`." } ], "toolCount": 10, "license": null, "provider": null, "endpoint": "https://mcp.finddatatech.cloud/fd-legal-search-mcp/mcp" }, { "id": "mcp:fd-paper-trading", "slug": "mcp-fd-paper-trading", "category": "mcp", "marketCategory": null, "tags": [], "canonical": { "brief": "柏讯模拟交易 MCP（paper-only）：多租户账户/订单/持仓 + 行情（概念族可配） + 策略双轨与回测。调用者身份逐请求委托（进站 Authorization 转发）。" }, "human": { "zh-CN": { "name": "模拟交易", "summary": "纸面交易的账户、下单与持仓，配行情读取与策略回放，不涉及真实资金。", "whenToUse": "验证交易策略或演练下单流程而不动真钱时。" }, "en": { "name": "Paper trading", "summary": "Accounts, orders, and positions on paper, with market reads and strategy replay; no real money involved.", "whenToUse": "To validate a strategy or rehearse order flow without real funds." }, "es": { "name": "Trading simulado", "summary": "Cuentas, órdenes y posiciones en papel, con lectura de mercado y repetición de estrategias; sin dinero real.", "whenToUse": "Para validar una estrategia o ensayar el flujo de órdenes sin fondos reales." }, "fr": { "name": "Trading fictif", "summary": "Comptes, ordres et positions sur papier, avec lecture de marché et rejeu de stratégies ; sans argent réel.", "whenToUse": "Pour valider une stratégie ou répéter le passage d'ordres sans fonds réels." }, "ja": { "name": "ペーパートレード", "summary": "紙上での口座・注文・建玉に加え、市況の読み取りと戦略の再生を提供します。実際の資金は使いません。", "whenToUse": "実資金を使わずに戦略を検証したり発注手順を練習するとき。" } }, "whenToUse": "验证交易策略或演练下单流程而不动真钱时。", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/mcp-catalog", "updatedAt": "2026-11-02" }, "tools": [ { "name": "place_order", "description": "Place a new paper trading order (MARKET or LIMIT)." }, { "name": "cancel_order", "description": "Cancel a pending or partially-filled order." }, { "name": "list_orders", "description": "List orders for an account, optionally filtered by status." }, { "name": "list_accounts", "description": "List all accounts for the authenticated user.\n\nReturns:\n List of accounts with IDs, names, and current cash balances.\n Empty list on error." }, { "name": "get_account", "description": "Get detailed information about a specific account." }, { "name": "create_account", "description": "Open a new paper account for the authenticated user.\n\nUse this when the user has no account yet (check with list_accounts first).\nAccounts start with the deployment's default starting balance when\nstarting_balance is omitted." }, { "name": "list_positions", "description": "List open positions for an account with quantity, average cost, and P&L." }, { "name": "search_instruments", "description": "Search the tradable instrument universe by symbol or company name." }, { "name": "get_bars", "description": "Get historical OHLCV bars for a symbol, oldest-first." }, { "name": "list_strategy_templates", "description": "List the built-in strategy templates with their parameter JSON Schemas.\n\nUse these to build a template-track strategy spec:\n{\"kind\": \"template\", \"template\": , \"symbol\": \"AAPL\", \"params\": {...}}.\n\nReturns:\n List of {id, name, description, params_schema}. Empty list on error." }, { "name": "create_strategy", "description": "Create a strategy from a spec." }, { "name": "update_strategy", "description": "Update a strategy. Changing the spec bumps its version; name/description\nedits do not." }, { "name": "list_strategies", "description": "List all strategies for the authenticated user.\n\nReturns:\n List of strategies with id, name, spec, spec_summary, and version.\n Empty list on error." }, { "name": "get_strategy", "description": "Get one strategy by id." }, { "name": "run_backtest", "description": "Start a backtest of a strategy over a daily-bar window.\n\nThe backtest runs in the background; poll get_strategy_run(run_id) until\nstatus is COMPLETED (stats + equity curve) or FAILED (error message)." }, { "name": "get_strategy_run", "description": "Get a strategy run's status, stats, and equity curve." }, { "name": "list_strategy_runs", "description": "List strategy runs (execution history), optionally filtered." }, { "name": "start_strategy", "description": "Start a live paper run of a strategy against a paper account.\n\nThe strategy's signals are evaluated on each new daily bar and turned\ninto orders on the account. Poll get_strategy_run for status\n(STARTING -> RUNNING; ERROR on failure)." }, { "name": "stop_strategy", "description": "Stop a live paper run. Cancels the run's resting LIMIT orders but never\ncloses its positions." } ], "toolCount": 19, "license": null, "provider": null, "endpoint": "http://cheap6.mesh.finddata.cn:31894/mcp/" }, { "id": "mcp:vaas-wanxing", "slug": "mcp-vaas-wanxing", "category": "mcp", "marketCategory": null, "tags": [ "vaas", "generation", "wanxing", "multimodal" ], "canonical": { "brief": "VAAS generation plane (Wanxing pilot): voice/image/video/cover generation, asset store, registry discovery. Multi-tenant; assets owner-scoped; publish plane excluded." }, "human": { "zh-CN": { "name": "萬星生成", "summary": "语音、图像、视频与封面的生成任务提交与状态查询，附资产库与模型音色目录。", "whenToUse": "需要生成配音、配图、短视频或封面并取回产物时。" }, "en": { "name": "Wanxing generation", "summary": "Submit and track voice, image, video, and cover generation jobs, with an asset store and model/voice catalogue.", "whenToUse": "When you need narration, stills, short video, or covers and want the produced assets back." }, "es": { "name": "Generación Wanxing", "summary": "Envío y seguimiento de trabajos de voz, imagen, vídeo y portada, con almacén de activos y catálogo de modelos y voces.", "whenToUse": "Cuando necesitas narración, imágenes, vídeo corto o portadas y quieres recuperar los activos." }, "fr": { "name": "Génération Wanxing", "summary": "Soumission et suivi de tâches voix, image, vidéo et couverture, avec un magasin d'actifs et un catalogue de modèles et de voix.", "whenToUse": "Pour produire une narration, des visuels, une vidéo courte ou une couverture et récupérer les actifs." }, "ja": { "name": "萬星ジェネレーション", "summary": "音声・画像・動画・カバーの生成ジョブを投入し、状態を確認。資産ストアとモデル・音声の一覧も提供します。", "whenToUse": "ナレーション、画像、短尺動画、カバーを生成して成果物を受け取りたいとき。" } }, "whenToUse": "需要生成配音、配图、短视频或封面并取回产物时。", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/mcp-catalog", "updatedAt": "2026-11-02" }, "tools": [ { "name": "generate_voice", "description": "Generate speech from text using TTS (registry method: tts).\n\nPass idempotency_key to make the call replay-safe: a second call with the\nsame key returns the existing asset (idempotent_replay=True) instead of\nre-generating and re-billing." }, { "name": "generate_image", "description": "Generate images from a text prompt (registry method: seedream)." }, { "name": "generate_cover", "description": "Generate a brand-consistent cover image via Remotion (registry method: brand)." }, { "name": "generate_video_submit", "description": "Submit an async video job; returns a job id in seconds (registry method: seedance).\n\nPoll video_job_status until succeeded/failed. Client timeouts may safely\nretry this call's sibling status tool — never resubmit." }, { "name": "video_job_status", "description": "Poll one async video job by id (running/succeeded/failed/expired).\n\nOn success the returned payload carries asset_id and a signed download_url." }, { "name": "models_voices", "description": "List available generation models and voices per provider (truth table).\n\nRead before choosing a voice/model: shows capabilities (speed/captions\nsupport), current status, and pricing notes per lane. The manifest at\ngenerators/voices_manifest.json is the source of truth for this tool." }, { "name": "voiceover_assemble", "description": "Assemble a long voiceover from a script plan (vaas-pipeline W2).\n\nsegments: ordered [{text, window_s?}]. Per-segment TTS with bounded retries;\nover-window segments get edit suggestions (never tempo-hacked). Deterministic\nplan slug; same plan + idempotency_key replays without re-billing." }, { "name": "reference_clips_check", "description": "Review reference clips for every active voice; regenerate missing/recycled.\n\nRun after upstream/model changes to produce a per-voice diff report." }, { "name": "video_compose", "description": "Compose one video from a storyboard, deterministically (vaas-pipeline W3).\n\nEvery frame is recomputed from time t — same storyboard → same byte stream.\nScene kinds: html_intro/html_outro (inline seek-contract page), kenburns\n(image list with slow pan/zoom), html_full (custom full-page HTML)." }, { "name": "orchestrate_voiceover", "description": "Create a complete voiceover video: TTS + captions + Remotion render." }, { "name": "orchestrate_brainstorm", "description": "Generate a structured content plan for voiceover or article." }, { "name": "assets_list", "description": "List asset summaries with optional filters (own assets only in hosted mode)." }, { "name": "assets_get", "description": "Get full asset details (lineage, variants, distribution); own assets only in hosted mode.\n\nIn hosted mode the response carries a fresh signed download_url so remote\ntenants can pull the artifact without a generation call." }, { "name": "assets_stats", "description": "Aggregate stats by asset type/stage and platform (own assets in hosted mode)." }, { "name": "assets_list_common", "description": "List common assets in downloads/common and remotion public." }, { "name": "assets_find_logo", "description": "Find the company logo asset(s)." }, { "name": "assets_list_compositions", "description": "List registered Remotion compositions (id + duration)." }, { "name": "assets_validate_paths", "description": "Validate staticFile() asset paths referenced in Remotion source." }, { "name": "assets_get_scene_templates", "description": "List scene template components from scenes*.tsx files." }, { "name": "registry_list_generators", "description": "List available generation methods, optionally filtered by asset type." }, { "name": "registry_describe_generator", "description": "Describe a specific generation method." }, { "name": "create_cover_image", "description": "Generate a brand-consistent cover image via Remotion (registry method: brand). (deprecated: Use generate_cover)" }, { "name": "create_voiceover", "description": "Create a complete voiceover video: TTS + captions + Remotion render. (deprecated: Use orchestrate_voiceover)" }, { "name": "brainstorm_content", "description": "Generate a structured content plan for voiceover or article. (deprecated: Use orchestrate_brainstorm)" }, { "name": "list_all_content", "description": "List asset summaries with optional filters (own assets only in hosted mode). (deprecated: Use assets_list)" }, { "name": "get_content_details", "description": "Get full asset details (lineage, variants, distribution); own assets only in hosted mode.\n\n In hosted mode the response carries a fresh signed download_url so remote\n tenants can pull the artifact without a generation call.\n (deprecated: Use assets_get)" } ], "toolCount": 26, "license": null, "provider": null, "endpoint": "https://mcp.finddatatech.cloud/vaas-wanxing/mcp" }, { "id": "mcp:wire-build-mcp", "slug": "mcp-wire-build-mcp", "category": "mcp", "marketCategory": null, "tags": [ "security-pending" ], "canonical": { "brief": "wire conversational build+steward" }, "human": { "zh-CN": { "name": "柏讯建造与治理", "summary": "柏讯的写入面：建数据集、建表单与记录视图、产物状态迁移、行更正与导出，全部落草稿。", "whenToUse": "需要新建数据集或表单、发布产物、更正数据行时。" }, "en": { "name": "BaiXun build & governance", "summary": "Write-side for BaiXun: create datasets, forms and record views, move artifacts through states, correct rows, and export — all starting as drafts.", "whenToUse": "When creating a dataset or form, publishing an artifact, or correcting rows." }, "es": { "name": "Construcción y gobierno BaiXun", "summary": "Lado de escritura en BaiXun: crear conjuntos de datos, formularios y vistas, cambiar estados de artefactos, corregir filas y exportar, siempre desde borrador.", "whenToUse": "Al crear un conjunto o formulario, publicar un artefacto o corregir filas." }, "fr": { "name": "Construction et gouvernance BaiXun", "summary": "Côté écriture de BaiXun : créer jeux de données, formulaires et vues, faire évoluer les artefacts, corriger des lignes et exporter, toujours à partir d'un brouillon.", "whenToUse": "Pour créer un jeu de données ou un formulaire, publier un artefact ou corriger des lignes." }, "ja": { "name": "柏訊ビルドとガバナンス", "summary": "柏訊の書き込み面：データセット・フォーム・レコードビューの作成、成果物の状態遷移、行の訂正とエクスポート。すべて下書きから始まります。", "whenToUse": "データセットやフォームの作成、成果物の公開、行の訂正を行うとき。" } }, "whenToUse": "需要新建数据集或表单、发布产物、更正数据行时。", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/mcp-catalog", "updatedAt": "2026-11-02" }, "tools": [ { "name": "wire_dataset_schema", "description": "wire dataset schema" }, { "name": "wire_artifact_list", "description": "wire artifact list" }, { "name": "wire_artifact_get", "description": "wire artifact get" }, { "name": "wire_dataset_create", "description": "wire dataset create" }, { "name": "wire_artifact_form_create", "description": "wire artifact form create" }, { "name": "wire_artifact_record_view_create", "description": "wire artifact record view create" }, { "name": "wire_artifact_update", "description": "wire artifact update" }, { "name": "wire_artifact_publish", "description": "wire artifact publish" }, { "name": "wire_artifact_archive", "description": "wire artifact archive" }, { "name": "wire_artifact_restore", "description": "wire artifact restore" }, { "name": "wire_dataset_rows_read", "description": "wire dataset rows read" }, { "name": "wire_dataset_correct", "description": "wire dataset correct" }, { "name": "wire_dataset_corrections_list", "description": "wire dataset corrections list" }, { "name": "wire_dataset_correction_revert", "description": "wire dataset correction revert" }, { "name": "wire_dataset_export", "description": "wire dataset export" }, { "name": "wire_artifact_dashboard_create", "description": "wire artifact dashboard create" }, { "name": "wire_artifact_doc_create", "description": "wire artifact doc create" } ], "toolCount": 17, "license": null, "provider": null, "endpoint": "http://cheap6.mesh.finddata.cn:31898/mcp" }, { "id": "mcp:wire-ops-mcp", "slug": "mcp-wire-ops-mcp", "category": "mcp", "marketCategory": null, "tags": [ "security-pending" ], "canonical": { "brief": "wire customer ops read face" }, "human": { "zh-CN": { "name": "柏讯客户操作", "summary": "柏讯客户面的读取操作：数据浏览与查询、需求单查看、市场订阅查看，以及内容供给的三项检索。", "whenToUse": "在柏讯里查数据、看需求单进度或浏览订阅内容时。" }, "en": { "name": "BaiXun customer operations", "summary": "Read-side operations for BaiXun customers: data browsing and queries, request lookup, subscription listing, plus three content-supply lookups.", "whenToUse": "When checking data, request progress, or subscribed content inside BaiXun." }, "es": { "name": "Operaciones de cliente BaiXun", "summary": "Operaciones de lectura para clientes BaiXun: navegación y consultas de datos, solicitudes, suscripciones y tres búsquedas de contenido.", "whenToUse": "Para consultar datos, el estado de una solicitud o el contenido suscrito en BaiXun." }, "fr": { "name": "Opérations client BaiXun", "summary": "Opérations de lecture pour les clients BaiXun : navigation et requêtes de données, demandes, abonnements et trois recherches de contenu.", "whenToUse": "Pour consulter des données, l'avancement d'une demande ou le contenu souscrit dans BaiXun." }, "ja": { "name": "柏訊カスタマー操作", "summary": "柏訊の顧客向け読み取り操作：データ閲覧・照会、依頼の確認、購読一覧、コンテンツ供給の検索三项です。", "whenToUse": "柏訊でデータ確認、依頼の進捗、購読コンテンツの閲覧をするとき。" } }, "whenToUse": "在柏讯里查数据、看需求单进度或浏览订阅内容时。", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/mcp-catalog", "updatedAt": "2026-11-02" }, "tools": [ { "name": "wire_dataset_aggregate", "description": "" }, { "name": "wire_dataset_query", "description": "" }, { "name": "wire_dataset_rows", "description": "" }, { "name": "wire_datasets_list", "description": "" }, { "name": "wire_demand_read", "description": "" }, { "name": "wire_demand_submit", "description": "" }, { "name": "wire_demands_list", "description": "" }, { "name": "wire_market_browse", "description": "" }, { "name": "wire_market_subscribe", "description": "" }, { "name": "wire_market_unsubscribe", "description": "" }, { "name": "wire_packs_list", "description": "" }, { "name": "wire_pack_subscribe", "description": "" }, { "name": "wire_pack_unsubscribe", "description": "" } ], "toolCount": 13, "license": null, "provider": null, "endpoint": "http://cheap6.mesh.finddata.cn:31896/mcp" }, { "id": "mcp:fd-wire-cell-drillcust001", "slug": "mcp-fd-wire-cell-drillcust001", "category": "mcp", "marketCategory": null, "tags": [], "canonical": { "brief": "fd-wire cell entry (managed by cell-manager)" }, "human": { "zh-CN": { "name": "柏讯 cell · drillcust001", "summary": "fd-wire cell entry (managed by cell-manager)" }, "en": { "name": "柏讯 cell · drillcust001", "summary": "fd-wire cell entry (managed by cell-manager)" }, "es": { "name": "柏讯 cell · drillcust001", "summary": "fd-wire cell entry (managed by cell-manager)" }, "fr": { "name": "柏讯 cell · drillcust001", "summary": "fd-wire cell entry (managed by cell-manager)" }, "ja": { "name": "柏讯 cell · drillcust001", "summary": "fd-wire cell entry (managed by cell-manager)" } }, "whenToUse": "fd-wire cell entry (managed by cell-manager)", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/mcp-catalog", "updatedAt": "2026-11-02" }, "tools": [ { "name": "alerts_list_series", "description": "Distinct (source, function_name, indicator) from observations + scraw_* tables,\neach with a row count + latest date. Alert-scoped discovery (not a SQL browser)." }, { "name": "alerts_get_series_latest", "description": "Return the latest `limit` rows of a series ordered by date_column desc." }, { "name": "alerts_create_alert_rule", "description": "Create a trigger rule over a DB series.\n\n`condition` is a safe DSL (no eval): `latest > 70`, `crosses_above(70)`,\n`pct_change(5) > 0.05`, `latest > 70 and prev <= 70`, etc. `channels` is a\nlist of names ([\"telegram\",\"slack\"]) or a {name: override} object. All\nchannel credentials live in root .env under ALERTS_*." }, { "name": "alerts_list_alert_rules", "description": "Return all alert rules." }, { "name": "alerts_get_alert_rule", "description": "Return one rule by name." }, { "name": "alerts_update_alert_rule", "description": "Update a rule's fields. Only provided fields change. The name cannot be renamed." }, { "name": "alerts_delete_alert_rule", "description": "Delete a rule. Its alert_events rows are removed via FK CASCADE." }, { "name": "alerts_list_channels", "description": "Return {name, configured, missing_keys} for each of the 7 channels\n(telegram, discord, slack, twitter, dingtalk, feishu, wecowork).\n\nNever returns credential values — only the names of missing keys." }, { "name": "alerts_run_alert_rule", "description": "Ad-hoc: evaluate one rule now and dispatch if it fires.\n\nRespects fire_mode + cooldown like the cron path. Use to test a rule on demand." }, { "name": "alerts_list_events", "description": "Recent alert_events (optionally for one rule), newest first." }, { "name": "cron_create_schedule", "description": "Create a new schedule with a cron expression and task." }, { "name": "cron_list_schedules", "description": "List all schedules." }, { "name": "cron_get_schedule", "description": "Get details for a single schedule." }, { "name": "cron_delete_schedule", "description": "Delete a schedule permanently." }, { "name": "cron_pause_schedule", "description": "Pause a schedule (disable without deleting)." }, { "name": "cron_resume_schedule", "description": "Resume a paused schedule." }, { "name": "cron_run_now", "description": "Execute a scheduled task immediately (one-shot)." }, { "name": "cron_list_executions", "description": "List execution history, optionally filtered by schedule_id." }, { "name": "cron_list_tasks", "description": "List available task types in the registry." }, { "name": "cron_create_task", "description": "Register a new task in the database. The command is a shell command or script path.\n\nExample: create_task(name=\"fetch_mofcom\", command=\"python3 scripts/fetch_mofcom_news.py --json\")" }, { "name": "cron_delete_task", "description": "Delete a task from the database by name." }, { "name": "cron_list_db_tasks", "description": "List all tasks stored in the database (user-defined, no code changes needed)." }, { "name": "cron_update_task", "description": "Update an existing DB task's command, description, or timeout. Only provided fields are updated." }, { "name": "composite_list", "description": "List all defined composites (name + description)." }, { "name": "composite_create", "description": "Create a new composite. Raises if the name already exists." }, { "name": "composite_list_upstreams", "description": "List the upstreams defined for a composite." }, { "name": "composite_add_upstream", "description": "Add an upstream MCP server to a composite.\n\nFor stdio: provide command, args, cwd (env optional).\nFor http: provide url." }, { "name": "composite_remove_upstream", "description": "Remove an upstream (and its selected tools) from a composite." }, { "name": "composite_list_available_tools", "description": "List the tools an upstream currently exposes (live), optionally filtered." }, { "name": "composite_add_tool", "description": "Select an upstream tool to be exposed (proxied) by the composite.\n\nThe tool is served as _ on next start.\n(alias is accepted but unused in v1.)" }, { "name": "composite_remove_tool", "description": "Remove a selected tool from a composite." }, { "name": "composite_list_tools", "description": "List the tools currently selected for a composite." }, { "name": "composite_add_chained_tool", "description": "Define a chained tool: a linear pipeline of upstream tool calls.\n\nEach step is {\"upstream\": , \"tool\": , \"input\": {param: value}}.\nInput values may be literals or references: \"$prev.\" (prior step's\nresult) or \"$step[N].\" (any prior step's result). Linear only — no\nbranching/conditionals." }, { "name": "composite_remove_chained_tool", "description": "Remove a chained tool from a composite." }, { "name": "composite_list_chained_tools", "description": "List the chained tools defined for a composite." }, { "name": "composite_create_manifest", "description": "Create a composite manifest: composite + upstreams + tools + workflows + prompt in one call." }, { "name": "composite_update_manifest", "description": "Partially update a composite manifest. Only provided fields change.\n\nupstreams/tools (when given) REPLACE the existing sets wholesale." }, { "name": "composite_delete_manifest", "description": "Delete a composite manifest (cascades upstreams/tools/chains)." }, { "name": "composite_list_manifests", "description": "List all composites as manifests (name, upstreams, tools, workflows, prompt)." }, { "name": "daas_list_sources", "description": "List all configured DAAS data sources with function counts.\n\nReturns sources like akshare, worldbank, ckan, cnstats\nwith install status and available function counts." }, { "name": "daas_search_functions", "description": "Search DAAS functions by name, category, or description across all sources." }, { "name": "daas_get_function_detail", "description": "Get full details for a DAAS function — parameters, output columns, description." }, { "name": "daas_list_categories", "description": "List all DAAS function categories with counts, optionally filtered by source." }, { "name": "daas_fetch_data", "description": "Execute a DAAS data function and return results as JSON.\n\nRoutes to the correct source adapter based on function name prefix." }, { "name": "daas_create_datasource", "description": "Create a new managed datasource." }, { "name": "daas_update_datasource", "description": "Update mutable fields of a datasource, identified by name or id.\n\nPass clear_category=true to unset the category (move to root level).\nPass score= to set the default score; pass clear_score=true to reset\nthe score back to NULL (unset). Only supplied fields are changed." }, { "name": "daas_delete_datasource", "description": "Delete a datasource and cascade-delete its forms, sections, and\ncollection-item references. Identify by name or id." }, { "name": "daas_create_category", "description": "Create a category, optionally under a parent (building the tree)." }, { "name": "daas_move_category", "description": "Re-parent a category. Rejects moves that would create a cycle\n(into own descendant) or self-parenting. Pass parent_id=null to make\nit a root." }, { "name": "daas_delete_category", "description": "Delete a category. Rejected if it has child categories. Datasources\nassigned to it are orphaned to root level (category_id set to null)." }, { "name": "daas_get_category_tree", "description": "Return the category tree (or a subtree from root_id) as a nested\nstructure, each node annotated with datasource_count." }, { "name": "daas_add_form", "description": "Add a form (e.g. '10-K', '8-K') to a datasource." }, { "name": "daas_add_section", "description": "Add a section to a form, carrying an extraction instruction." }, { "name": "daas_list_forms", "description": "List all forms of a datasource, each with nested sections\n(including instruction)." }, { "name": "daas_create_collection", "description": "Create a named datasource collection." }, { "name": "daas_add_to_collection", "description": "Add a datasource (or a specific datasource-section) to a collection.\n\nOmit section_name to add the whole datasource; supply section_name to\nadd only that section. Optional score sets a per-collection score override\non the new item (NULL = inherit the datasource's default score)." }, { "name": "daas_list_collection_items", "description": "List a collection's items, each resolved to source name and (if set)\nsection name + instruction." }, { "name": "daas_remove_from_collection", "description": "Remove an item from a collection by (collection, source, optional\nsection). Omit section_name to remove the whole-datasource item." }, { "name": "daas_list_collections", "description": "List every datasource collection with its item count." }, { "name": "daas_rename_collection", "description": "Rename a collection. The new name must be unique. Items are preserved\n(they reference by collection_id)." }, { "name": "daas_update_collection", "description": "Partially update a collection's name and/or description.\n\nAt least one of `new_name` / `description` must be provided; omitted\nfields are left unchanged. When `new_name` differs from the current\nname it must be unique. Raises if the collection is not found." }, { "name": "daas_delete_collection", "description": "Delete a collection. Cascades to its `datasource_collection_items`\nrows (datasources themselves are untouched)." }, { "name": "daas_reorder_collection_items", "description": "Rewrite the sort_order of items in `collection_name` to match the\ngiven list. `ordered_item_ids` MUST include every existing item id in\nthe collection exactly once — partial reorders are rejected." }, { "name": "daas_set_collection_item_score", "description": "Set or clear the per-collection score override on an existing collection\nitem, identified by (collection_name, source_name, optional section_name).\n\nPass score= to set the override; pass score=null (omit) to clear it\nso the item falls back to the datasource's default score. Returns the\nupdated item with its resolved effective score, the raw item_score override,\nand the source_default_score." }, { "name": "daas_search_datasources", "description": "Search/filter datasources across levels — category (with optional\nsubtree), source, form, and section — every filter optional, plus a\nfree-text query across source label/description, form label, and\nsection name/instruction." }, { "name": "daas_search_entities", "description": "Search entities (stocks + countries) by name, ticker, code, or alias." }, { "name": "daas_get_entity", "description": "Get full detail for one entity, including aliases and metadata.\n\nPer D5/3.5 the entity master is fd-open-data-mcp. Resolve by natural\nkey (entity_type, code) when given (gateway-first, local fallback);\notherwise fall back to entity_id (local)." }, { "name": "daas_list_entities", "description": "List entities filtered by type / exchange / country, paginated." }, { "name": "daas_get_entity_coverage", "description": "For each datasource linked to the entity, return the identifier to\nuse, the available sections (routing instructions = how to get the\ndata, with an identifier-prefilled variant), and the column count/list\naggregated from daas_function_columns. Sources without registered\nfunctions get a column_hint naming the sibling MCP + tool.\n\nPer D5/3.5 entity resolution proxies through fd-open-data-mcp. Accept\na natural key (entity_type, code) as an alternative to entity_id.\n\nThis answers: \"I have company X — which datasources cover it, how many\ncolumns can I get, and how do I fetch it?\"" }, { "name": "daas_link_entity_datasource", "description": "Link an entity to a daas datasource, recording the identifier to use\ninside that datasource (e.g. for AAPL → yfinance: 'AAPL'). Upserts on\n(entity_id, source_name)." }, { "name": "daas_unlink_entity_datasource", "description": "Delete the link between an entity and a datasource." }, { "name": "daas_create_entity_collection", "description": "Create a named entity collection (watchlist / portfolio)." }, { "name": "daas_list_entity_collections", "description": "List every entity collection with its item count." }, { "name": "daas_get_entity_collection", "description": "Get one collection with its current member entities (ordered)." }, { "name": "daas_update_entity_collection", "description": "Partially update an entity collection's name and/or description and/or\nrule. At least one field must be provided. `clear_rule=True` resets the\nrule to NULL (manual collection)." }, { "name": "daas_delete_entity_collection", "description": "Delete a collection and cascade to its items + changes." }, { "name": "daas_add_entity_to_collection", "description": "Add an entity to a collection. Resolves the entity by `entity_id`, or\nby `(entity_type, code)`. Records an `add_in` event in\n`entity_collection_changes` (source='manual'). No-op (no event recorded)\nif the entity is already a member." }, { "name": "daas_remove_entity_from_collection", "description": "Remove an entity from a collection. Records a `remove_out` event in\n`entity_collection_changes` (source='manual'). No-op (no event recorded)\nif the entity is not a member." }, { "name": "daas_list_entity_collection_items", "description": "List the current members of a collection, ordered by sort_order, each\nenriched with full entity detail (code, name, ticker, exchange, ...)." }, { "name": "daas_reorder_entity_collection_items", "description": "Rewrite member sort_order to match the given ordered list of\n`entity_collection_items.id`. Must contain exactly the current item ids." }, { "name": "daas_list_entity_collection_changes", "description": "Query the add-in / remove-out audit log (`entity_collection_changes`),\nnewest first, each enriched with collection name + entity code/name." }, { "name": "daas_sync_entity_collection", "description": "Re-derive the member set for a rule-based collection by applying its\n`rule_json` to the `entities` table, diff vs current members, apply\nadd_in for new matches and remove_out for non-matches (source='cron'),\nand record every transition. Returns `{added, removed, unchanged}`.\nA manual collection (rule_json=NULL) is a no-op." }, { "name": "daas_create_indicator_collection", "description": "Create a named indicator collection (a reusable bundle of indicators)." }, { "name": "daas_list_indicator_collections", "description": "List every indicator collection with its item count." }, { "name": "daas_get_indicator_collection", "description": "Get one collection with its current items (ordered, each with the\nresolved effective score + raw item/indicator/datasource scores)." }, { "name": "daas_update_indicator_collection", "description": "Partially update an indicator collection's name, description, and/or\nrule. At least one field must be provided. `clear_rule=True` resets\n`rule_id` to NULL (manual collection)." }, { "name": "daas_delete_indicator_collection", "description": "Delete a collection and cascade to its items + changes." }, { "name": "daas_add_indicator_to_collection", "description": "Add an indicator to a collection. Resolves the indicator by `name`." }, { "name": "daas_remove_indicator_from_collection", "description": "Remove an indicator from a collection. Records a `remove_out` event in\nthe audit log. No-op (`not_member`) if the indicator is not in the collection." }, { "name": "daas_list_indicator_collection_items", "description": "List the items of a collection, ordered by sort_order, each with the\nresolved effective score + raw item/indicator/datasource scores." }, { "name": "daas_reorder_indicator_collection_items", "description": "Rewrite item sort_order to match the given ordered list of\n`indicator_collection_items.id`. Must contain exactly the current item\nids of this collection — partial reorders are rejected." }, { "name": "daas_set_indicator_collection_item_score", "description": "Set or clear the per-collection score override on an existing item,\nidentified by (collection_name, indicator_name).\n\n`score` = a float sets the override; `score = null` clears it (the item\nthen inherits the indicator's default `indicator_rules.score`, which\nitself inherits the datasource default when NULL). Returns the updated\nitem dict with the resolved effective score." }, { "name": "daas_list_indicator_collection_changes", "description": "Query the add-in / remove-out audit log (`indicator_collection_changes`),\nnewest first. Each row is enriched with the collection name." }, { "name": "daas_sync_indicator_collection", "description": "Re-derive the member set for a rule-based indicator collection by\nevaluating its rule (target='indicator_names') via the RuleEngine, diff vs\ncurrent members, apply add_in/remove_out (source='cron'), and record every\ntransition. Returns `{added, removed, unchanged}`. A manual collection\n(rule_id NULL) is a no-op." }, { "name": "daas_create_pipeline_collection", "description": "Create a named pipeline collection." }, { "name": "daas_list_pipeline_collections", "description": "Every pipeline collection with its item count." }, { "name": "daas_get_pipeline_collection", "description": "One pipeline collection with all its items." }, { "name": "daas_delete_pipeline_collection", "description": "Unwire cron for every item, then delete the collection (FK CASCADE removes items)." }, { "name": "daas_list_pipeline_items", "description": "List items, optionally filtered to one collection." }, { "name": "daas_add_pipeline_item", "description": "Validate + store an item. If enabled and backfill, run history backfill\nthen register the cron-mcp schedule. Returns {item, backfill, cron}." }, { "name": "daas_remove_pipeline_item", "description": "Unwire cron then delete the item row. scraw_
is left intact." }, { "name": "daas_enable_pipeline_item", "description": "Backfill + register cron for a disabled item; set enabled=1." }, { "name": "daas_disable_pipeline_item", "description": "Delete the cron schedule (keep the task row); set enabled=0." }, { "name": "daas_update_pipeline_item", "description": "Update mutable fields. cron/timezone change re-syncs the schedule.\nenabled toggle runs enable/disable. arguments_json change does NOT backfill." }, { "name": "daas_sync_pipeline_cron", "description": "Re-apply cron wiring for all enabled items; remove schedules for disabled ones." }, { "name": "daas_list_models", "description": "Return configured models with a `vision` flag each. API keys are never returned." }, { "name": "daas_list_source_tables", "description": "Return scraped source-data tables (name LIKE 'scraw_%') with row counts + columns." }, { "name": "daas_create_rule", "description": "Create a rule." }, { "name": "daas_list_rules", "description": "List all rules, optionally filtered by rule_type." }, { "name": "daas_get_rule", "description": "Return one rule by name." }, { "name": "daas_update_rule", "description": "Partially update a rule. At least one field must be provided. The rule\nname cannot be renamed." }, { "name": "daas_delete_rule", "description": "Delete a rule. Referencing collections' rule_id is set to NULL; the\nrule's process_results rows are removed." }, { "name": "daas_run_rule", "description": "Evaluate + persist. For `target='rows'` (llm): incremental extraction\ninto process_results + advance last_rowid. For member targets: returns the\nderived set (membership is applied by daas_sync_*_collection, not here)." }, { "name": "daas_test_rule", "description": "Dry-run: evaluate the rule WITHOUT persisting. For member targets\n(entity_ids/indicator_names) returns the derived set. For `target='rows'`\n(llm), extracts from a single source row (no process_results write) as a\nsample." }, { "name": "daas_extract_text", "description": "Extract structured records from (possibly long) text.\n\nLong text is chunked (no truncation): per-chunk extraction then a merge pass." }, { "name": "daas_extract_image", "description": "Extract structured records from an image via a vision-capable model.\n\n`image` accepts a local file path, an http(s):// URL, or raw base64." }, { "name": "daas_extract_file", "description": "Read a local .txt/.md/.pdf and extract structured records." }, { "name": "daas_list_indicator_ops", "description": "Return the fixed math-op catalog with each op's required params." }, { "name": "daas_create_indicator", "description": "Create an indicator rule bound to a datasource + source table." }, { "name": "daas_list_indicators", "description": "Return all indicator rules (each with `score` + `effective_default_score`)." }, { "name": "daas_get_indicator", "description": "Return one indicator rule by name (with `score` + `effective_default_score`)." }, { "name": "daas_update_indicator", "description": "Update an indicator rule's fields. Only provided fields change. The\nrule name cannot be renamed. Pass `params` to replace the op params.\n\n`score`: a float sets the indicator's default score. `clear_score=True`\nclears it back to NULL (inherit the datasource's `sources.score`).\nMirrors `update_datasource(score=…, clear_score=True)`." }, { "name": "daas_set_indicator_score", "description": "Set the indicator's default `score` (float), or clear it (None → inherit\nthe datasource's `sources.score`). Returns the updated indicator dict with\n`effective_default_score`." }, { "name": "daas_delete_indicator", "description": "Delete an indicator rule. observations rows it produced are NOT removed\n(soft reference) — they remain identifiable via their `metadata.rule_name`." }, { "name": "daas_run_indicator", "description": "Full-recompute the indicator over its source table → observations.\n\nReads the whole source table, computes the op, and upserts every\n(date, value) into the daas `observations` table keyed on\n(source=datasource, function_name, indicator=indicator_name, date).\nIdempotent on re-run via the observations unique constraint." }, { "name": "daas_calculate_indicator", "description": "Ad-hoc: compute an indicator over a source table without persisting.\n\nValidates table/columns/op/params, reads the series, computes, and returns\n{indicator, dates, values, count}. Writes nothing (no rule, no observations)." }, { "name": "dashboard_list_databases", "description": "List all known SQLite databases in the MCP directory with table counts." }, { "name": "dashboard_query_table", "description": "Run a read-only query against a database table. Returns rows as JSON." }, { "name": "dashboard_list_datasources", "description": "List managed datasources from daas.db (schema managed by mcp/models/models.py)." }, { "name": "dashboard_get_executions", "description": "Get cron execution history from a database." }, { "name": "dashboard_get_stats", "description": "Get aggregated statistics across all MCP databases." }, { "name": "dashboard_register", "description": "Register or update (upsert by slug) a standalone HTML dashboard in the\n`dashboards` table. After writing the row, regenerates index.html + daas.md\nfrom the DB. `source_tables`/`entity_coverage`/`time_range`/`chart_config`\naccept a JSON string or a list/dict. Returns the stored row (with `action`\n= inserted|updated) or `{\"error\": ...}`." }, { "name": "dashboard_list", "description": "List every registered dashboard (name + slug + intro + file_url), oldest\nfirst. Returns a JSON array." }, { "name": "dashboard_get", "description": "Get one dashboard's full metadata by slug: name, intro, source_tables,\nentity_coverage, time_range, refresh_cadence, chart_config, file_path,\nfile_url. Returns `{\"error\": \"dashboard '' not found\"}` if absent." }, { "name": "dashboard_search", "description": "Case-insensitive keyword search over each dashboard's name + intro +\nrefresh_cadence + source_tables. Returns matching `{slug, name, intro}`\nentries, or an empty array if none match." }, { "name": "dashboard_update", "description": "Patch one or more fields of an existing dashboard (by slug). Only the\nfields you pass are changed; the rest are left as-is. Regenerates index.html\n+ daas.md after the update. Returns the updated row or `{\"error\": ...}`." }, { "name": "dashboard_delete", "description": "Delete a dashboard row by slug and regenerate index.html + daas.md. The\nHTML file itself is NOT removed (only the registry row). Returns\n`{\"deleted\": slug}` or `{\"error\": ...}`." }, { "name": "gateway_list_data_mcps", "description": "List the data-fetch MCP upstreams gateway-mcp can route to.\n\nEach entry is a row from the `gateway_upstreams` table (name, transport,\nenabled, description). Disabled upstreams are hidden unless\n`include_disabled=True`." }, { "name": "gateway_list_data_mcp_tools", "description": "List the tools exposed by a data-fetch MCP upstream (live).\n\nReuses the persistent client from the connection pool — the client stays\nalive across calls and is only recreated when the upstream config changes.\nReturns each tool's name + description (+ parameters when available)." }, { "name": "gateway_call_data_mcp", "description": "Call a tool on a data-fetch MCP upstream and return its result.\n\nConnects to the named upstream via the persistent client pool, invokes\nthe named tool with the JSON-deserialized `arguments`, and returns the\nupstream's result. The client stays alive across calls and is only\nrecreated when the upstream config changes." }, { "name": "gateway_add_data_mcp", "description": "Add or update a data-fetch MCP upstream (idempotent upsert by name)." }, { "name": "gateway_remove_data_mcp", "description": "Remove a data-fetch MCP upstream from the registry." }, { "name": "gateway_get_data_mcp", "description": "Get one data-fetch MCP upstream's full launch config." }, { "name": "gateway_gateway_health", "description": "Health-probe each enabled gateway upstream and auto-flip transport.\n\nFor each enabled upstream in ``gateway_upstreams``:\n\n- **http** transport: ping the row's ``url``. On failure AND the row\n carries a stdio ``command``, flip ``transport`` to ``stdio`` (degraded\n mode — mirrors the client-pool fallback so the next ``get_client()``\n launches the subprocess). On success, no change.\n- **stdio** transport with a ``url`` set (i.e. it was flipped from http by\n a prior fallback): ping the ``url``. On success flip ``transport`` back\n to ``http`` (recovered). On failure, leave as stdio.\n- **stdio**-only (no ``url``): skipped (nothing to probe).\n\nReturns a per-upstream status report. Never raises — network errors are\ncaptured as degraded states so the probe is safe to call from selfcheck." }, { "name": "research_create", "description": "Create a persisted research bundle. Named collections are attached by\nname; with ``create_missing=True`` missing entity/indicator/pipeline\ncollections are created empty. A ``dashboard_slug`` must already exist\n(the tool does not build dashboard HTML). ``component_refs`` is a JSON\nobject string: ``{\"rules\": [...], \"scraw_tables\": [...], \"indicators\": [...]}``." }, { "name": "research_get", "description": "Return a research + resolved bundle (live counts, dashboard url,\npipeline status, dangling-component warnings)." }, { "name": "research_list", "description": "List every research (optionally filtered by status) with attached\ncomponent names." }, { "name": "research_update", "description": "Partially update a research. ``None`` means leave a field unchanged;\nto detach a named component use ``remove_component``. Any newly-set\ncomponent name is validated to exist. ``component_refs`` (JSON string)\nreplaces the auxiliary refs." }, { "name": "research_delete", "description": "Delete a research. Removes the on-disk report file. With\n``remove_pipeline=True`` (default) deletes the owned pipeline collection\n+ orphaned cron schedules/tasks. With ``remove_dashboard=True`` (default\nFalse) deletes the referenced dashboard via the sibling tool. Shared\nentity/indicator collections, rules, scraw tables, and observations are\nNEVER deleted." }, { "name": "research_generate_report", "description": "Assemble a markdown report from the bundle's live components, persist it\ninto ``report_md`` and write it to ``researches/.md`` (dir from\n``RESEARCH_DIR``, default repo-root ``researches/``). Regeneration\noverwrites both. Returns ``{name, report_path, char_count}``." }, { "name": "research_refresh", "description": "Re-run the research's data pipeline: recompute each indicator in the\nindicator collection (via ``daas_run_indicator``), sync rule-based\ncollections (via ``daas_sync_entity_collection``/``daas_sync_indicator_collection``),\nand report each pipeline item's current status. Returns a per-component\nstatus report; per-item errors are reported, not raised." }, { "name": "research_add_component", "description": "Attach an existing component to a research by name. ``component_type`` ∈\nentity_collection | indicator_collection | dashboard | pipeline | rule |\nscraw_table | indicator. The underlying object is not created - it must\nalready exist (for the named kinds)." }, { "name": "research_remove_component", "description": "Detach a component from a research. The underlying object is NOT\ndeleted. For named kinds, clears the research column; for auxiliary refs,\nremoves the name from ``component_refs``." }, { "name": "workflow_register", "description": "Register a new workflow manifest (YAML). Version bumps on name collision\nunless the manifest declares its own ``version``." }, { "name": "workflow_get", "description": "Get a workflow by name (latest version by default)." }, { "name": "workflow_list", "description": "List every workflow (name, version, description, enabled, step_count)." }, { "name": "workflow_update", "description": "Patch a workflow. Only provided fields change. ``version`` defaults to\nthe latest version of ``name``." }, { "name": "workflow_delete", "description": "Delete a workflow. Omit ``version`` to delete every version of ``name``.\nSteps/runs cascade via FK." }, { "name": "workflow_run", "description": "Run a workflow manifest. ``params_json`` merges over the manifest's\ndeclared ``params``. Returns run id + status; a paused run carries a\n``resume_token`` for ``resume``." }, { "name": "workflow_inspect", "description": "Validate a stored manifest and return its parsed step plan (no execution)." }, { "name": "workflow_resume", "description": "Resume a paused run. ``approved=False`` marks it failed." }, { "name": "customer_dataset_ingest_begin", "description": "Begin a full-snapshot ingest for a customer dataset (registers/refreshes\nthe dataset metadata and creates the staging table). Supersedes any\nin-flight ingest for the same key." }, { "name": "customer_dataset_ingest_chunk", "description": "Append one chunk of rows to an open ingest. `start_index` must equal the\nnumber of rows already staged (guards against replays)." }, { "name": "customer_dataset_ingest_commit", "description": "Commit an open ingest per its begin-time mode: `replace` atomically\nswaps the staged snapshot into the base table; `merge` upserts the staged\nrows into the base by the dataset's primary key (rows absent from staging\nare kept). Either way the effective table is rebuilt and all active\ncorrections replayed (dangling anchors are marked, never dropped). Returns\nrows/size/corrections stats." }, { "name": "customer_dataset_ingest_abort", "description": "Abort an open ingest and drop its staging table (base/effective stay)." }, { "name": "customer_dataset_overview", "description": "Dataset metadata: columns, pk_columns, rows, size_bytes, last_commit_at.\nWhen FD_CELL_STORAGE_QUOTA_MB is armed on the PostgreSQL backend, also\nstorage_quota_bytes / storage_used_bytes (pg_database_size)." }, { "name": "customer_dataset_preview", "description": "Paged row preview with `__row_id` anchors. `include_corrected=false`\nreads the raw snapshot (original values stay queryable)." }, { "name": "customer_dataset_query", "description": "Structured query over a customer dataset: whitelisted filter operators\n(eq/ne/gt/gte/lt/lte/in/notin/like/isnull/notnull, AND-composed), multi-column\nsort, limit/offset paging, exact total and an honest truncated flag; rows\ncarry `__row_id` anchors. `count_only=True` returns just the count;\n`distinct_column` returns that column's distinct values (capped at 1,000);\n`raw=True` reads the base snapshot without edits. Responses are capped at\n1,000 rows and queries abort past 5s." }, { "name": "customer_dataset_aggregate", "description": "GROUP BY aggregation over a customer dataset: multi-column grouping ×\ncount/sum/avg/min/max measures, filter-compatible, group rows capped at\n1,000 with exact group_count. sum/avg reject TEXT columns (honest type\nguard); min/max allow any column. `raw=True` reads the base snapshot." }, { "name": "customer_dataset_correction_add", "description": "Add an overlay correction (update/delete/insert); it applies to the\neffective table immediately and is replayed after every future commit.\nEvery edit record carries its write channel (console | data-key | agent)\nfor audit attribution." }, { "name": "customer_dataset_correction_list", "description": "Correction history (audit trail) for a dataset." }, { "name": "customer_dataset_correction_dangling", "description": "Corrections whose row anchor no longer matches the latest snapshot." }, { "name": "customer_dataset_correction_revert", "description": "Revert one correction (mark reverted; rebuild the effective table from\nthe base snapshot plus the remaining active corrections)." }, { "name": "customer_dataset_write_batch", "description": "Business-write surface (wire-customer-data-engine §2): a list of edit\noperations applied in ONE transaction (≤1,000 ops). Any rejected op rolls\nthe whole batch back — zero partial writes — and the error carries the\n1-based index of the failing entry. Anchors are strict here: an\nupdate/delete that matches no row rejects the batch (business writes must\nhit their anchors; dangling is a connector-world outcome, not an app one)." }, { "name": "customer_dataset_create_declared", "description": "Schema-declared dataset creation: an EMPTY dataset from a column/type\ndeclaration (optionally with primary-key columns that become the row\nanchor), indistinguishable from an uploaded/connector-created dataset.\nThe app-first creation channel for business systems." }, { "name": "customer_dataset_add_column", "description": "Add ONE nullable column to a dataset (effective + base tables and the\nmetadata). Additive-only schema evolution: destructive changes are not\noffered. Rejected while an ingest is open." }, { "name": "customer_dataset_derive", "description": "Materialize a structured-query result over `parent_key` into a NEW\nchild dataset, lineage recorded (parent + spec). The child inherits the\nparent's columns and pk declaration. Re-derivation of an existing child\ngoes through `derive_refresh`." }, { "name": "customer_dataset_join_indicators", "description": "Derive a child dataset = customer rows × platform indicator series.\nSeries are fetched through the wire entrance with the cell's credential\n(metered to this customer, same track as artifact materialization).\nMatching: exact date → month period → year period (frequency inferred\nfrom the series' own points); misses join as NULL — never neighbor\nvalues. `region_column` (optional) resolves per-row entities by name;\nwithout it every row uses `default_entity`. At most 10 indicator codes\nper derivation and 64 distinct regions." }, { "name": "customer_dataset_link", "description": "Materialize a semantic connection between two customer datasets into a\nNEW child dataset (wire-entity-link): LEFT JOIN semantics — every left row\nis kept; rows whose key has no right counterpart carry NULL in the right\ncolumns (never a neighbor's value). Right columns whose name collides with\na left column (or an already-chosen right output name) are renamed\n``right_`` (then ``right__2`` …); uncolliding right columns\nkeep their names and types. The child inherits the LEFT dataset's pk\ndeclaration and its lineage is recorded as kind=\"link\" (parent = left\ndataset; the right dataset and both join columns live in the spec), so\n``derive_refresh`` re-joins both sides' latest effective state. The join\nruns on the effective tables (corrections apply) using SQL ``IS`` key\nequality — NULL keys compare equal to NULL, keeping equality semantics\nconsistent. Row counts beyond the cap reject the whole operation (the\nchild keeps its previous state).\n\nm:n 桥展开（wire-entity-mn-materialize）：给齐 junction 三要素\n（junction_key / junction_left_column / junction_right_column）时改走\n桥行驱动 INNER 对表——一条桥行一行结果、无桥关联端点行不入结果、\n中间集列不进输出、孤儿桥以 ``orphan_dropped`` 计数如实上报；三要素\n必须同给同缺。缺省时 n:1/1:1 行为逐字节等价。" }, { "name": "customer_dataset_derive_refresh", "description": "Re-run a derived dataset's stored spec against its parent's current\neffective state. Join derivations re-fetch their indicator series first\n(metered reads again); link derivations re-join both sides from the\nrecorded spec; operator (filter/project/aggregate/union) and SQL\nderivations replay their archived spec verbatim." }, { "name": "customer_dataset_lineage", "description": "Lineage of a dataset: its own derivation (parent + spec + last run)\nand its live derived children." }, { "name": "customer_dataset_delete", "description": "Delete a dataset: drop its three tables, metadata, corrections and\nsoft-reference rows. Ingest audit rows are kept." }, { "name": "customer_dataset_filter", "description": "Materialize a filtered (optionally sorted/limited) subset of\n`parent_key` into a NEW child dataset. Same filter/sort/limit semantics\nas the query surface and as `derive` — the difference is lineage kind\n`filter`, so pipelines and the dataset graph label this step truthfully\ninstead of calling it a generic derivation." }, { "name": "customer_dataset_project", "description": "Materialize a column projection (with renaming) into a NEW child\ndataset. Output column order follows the declaration order. Primary-key\ncolumns that are projected follow their new names; keys left out of the\nprojection are dropped (row anchoring then falls back to the row hash)." }, { "name": "customer_dataset_aggregate_derive", "description": "Materialize a GROUP BY result into a NEW child dataset (rows = groups,\ncolumns = group keys + measures). This is the materializing counterpart of\nthe read-only `aggregate` tool: same validation and type guards, but the\noutput becomes a real dataset that downstream steps can build on.\nsum/avg reject TEXT columns (honest type guard); min/max allow any column." }, { "name": "customer_dataset_union", "description": "Append two datasets vertically into a NEW child dataset, aligning\ncolumns by name: column set = union of both sides (first-seen order),\nmissing columns become NULL. Same-name columns with incompatible types\n(TEXT vs numeric) are rejected honestly — never silently coerced;\nINTEGER + REAL widens to REAL. The output has no primary key (append\nsemantics)." }, { "name": "customer_dataset_window", "description": "Materialize a partitioned/ordered window computation over `parent_key`\ninto a NEW child dataset: source columns passed through unchanged plus one\ncolumn per `window` row (named by `as`). The row set is unchanged (windows\nnever add or drop rows); lineage kind `window` records the full spec so\n`derive_refresh` replays it.\n\n函数闭集（design D1）：row_number/rank/dense_rank 与 lag/lead 必须有\n`order_by`（否则 422 指名函数）；lag/lead 的 `offset` 缺省 1、必须\n1..10000；sum/avg/min/max 的 `column` 必填且 ⊆ 列快照，count 允许无\ncolumn（全行计数）。`as` 必须给出、唯一、且不与源列名冲突（422 指名，\n不静默改名）。`frame ∈ whole_partition|running`（running 必须有\norder_by；闭集，不开放任意边界语法）；frame 只对聚合类生效——行级函数\n（row_number/rank/dense_rank/lag/lead）的取值与帧无关。\n\n确定性（design D2）：内部 ORDER BY 末尾追加稳定行键 `__row_hash` 作\n最终 tie-break，并列数据重派生逐行一致；`__row_hash` 与一切保留前缀列\n绝不进输出列。" }, { "name": "customer_dataset_sql", "description": "Materialize a read-only SQL query into a NEW child dataset. The SQL\ntext, placeholder mapping and parameters are archived in lineage\n(append-only) so the step can be replayed and audited; replay re-executes\nthe archived statement verbatim.\n\nRead-only by construction: single statement starting with SELECT/WITH,\nno DDL/DML/PRAGMA/ATTACH, dangerous functions denied, a sqlite authorizer\nthat denies anything but reads of the declared datasets, and the\nconnection itself opened in read-only mode. Results over the dataset row\nlimit reject the whole batch." }, { "name": "customer_dataset_sql_preview", "description": "Dry-run a read-only SQL step: returns at most 100 preview rows plus\nthe result-set schema and creates/refreshes NOTHING. Guardrails are\nidentical to `sql_derive` (single read-only statement, declared\nplaceholders only, literal parameter binding, dangerous functions\ndenied, read-only connection, timeout and row caps)." }, { "name": "customer_dataset_provenance_wal_pending", "description": "Read un-forwarded WAL events for the platform-side provenance\nforwarder: rows with ``acked = 0`` and ``seq > after_seq``, ascending by\n``seq`` (cursor style — hand the returned ``last_seq`` to\n``provenance_wal_ack``, then keep polling with it as ``after_seq``).\n``limit`` is clamped to [1, 1000]. Events stay readable until acked, so a\nforwarder crash between read and ack loses nothing. On a database that\npredates the WAL (table not yet created) returns an empty page." }, { "name": "customer_dataset_provenance_wal_ack", "description": "Ack WAL events up to and including ``upto_seq`` (forwarder confirms\nthey are durably in the platform ledger). Idempotent: re-acking returns\n``{\"acked\": 0}``. Flipping the ``acked`` bit is the only mutation the WAL\never allows — tamper-proofing lives at the platform ledger, not here." }, { "name": "customer_dataset_check", "description": "Batch content-level checks over a customer dataset\n(openspec: wire-quality-rules-v2). Each check: {kind, ...}:\n- regex: {kind: \"regex\", column, pattern} — pattern length ≤128, no\n nested quantifiers, executed with a deterministic timeout (regex lib);\n- col_compare: {kind: \"col_compare\", col_a, op, col_b} — ops:\n eq/ne/gt/gte/lt/lte;\n- range: {kind: \"range\", column, min?/max?} — closed interval, at least\n one bound.\n\nAt most 5 checks per call. Returns per-check Great-Expectations-style\noutcome {status: pass|fail|error, element_count, unexpected_count,\npass_rate}. NULL values never count as violations (null coverage is a\nseparate concern); an empty dataset passes vacuously. A check whose own\nexecution fails (e.g. regex timeout) reports `error`, never `fail`.\nFailure samples (offending values / row indexes) are aggregated to\ncounts inside the engine and discarded — the response contains no data\nvalues." }, { "name": "customer_dataset_quality", "description": "Quality snapshot aggregates for one dataset (openspec:\nwire-quality-rules-v2; payload shape aligned field-by-field with the\nplatform-side sanitizer). Returns {row_count, computed_at, pk_declared,\npk_uniqueness: {columns, unique_ratio} | null, null_rates: [{column,\nnull_rate, sampled_rows}], sample_limit_hit}. Only aggregate counts and\nratios — never data values." } ], "toolCount": 193, "license": null, "provider": null, "endpoint": "https://mcp.finddatatech.cloud/fd-wire-cell-drillcust001/mcp" }, { "id": "mcp:fd-wire-cell-pgauto001", "slug": "mcp-fd-wire-cell-pgauto001", "category": "mcp", "marketCategory": null, "tags": [], "canonical": { "brief": "fd-wire cell entry (managed by cell-manager)" }, "human": { "zh-CN": { "name": "柏讯 cell · pgauto001", "summary": "fd-wire cell entry (managed by cell-manager)" }, "en": { "name": "柏讯 cell · pgauto001", "summary": "fd-wire cell entry (managed by cell-manager)" }, "es": { "name": "柏讯 cell · pgauto001", "summary": "fd-wire cell entry (managed by cell-manager)" }, "fr": { "name": "柏讯 cell · pgauto001", "summary": "fd-wire cell entry (managed by cell-manager)" }, "ja": { "name": "柏讯 cell · pgauto001", "summary": "fd-wire cell entry (managed by cell-manager)" } }, "whenToUse": "fd-wire cell entry (managed by cell-manager)", "badges": { "official": false, "facade": false }, "provenance": { "source": "https://facet.finddatatech.cloud/api/mcp-catalog", "updatedAt": "2026-11-02" }, "tools": [ { "name": "alerts_list_series", "description": "Distinct (source, function_name, indicator) from observations + scraw_* tables,\neach with a row count + latest date. Alert-scoped discovery (not a SQL browser)." }, { "name": "alerts_get_series_latest", "description": "Return the latest `limit` rows of a series ordered by date_column desc." }, { "name": "alerts_create_alert_rule", "description": "Create a trigger rule over a DB series.\n\n`condition` is a safe DSL (no eval): `latest > 70`, `crosses_above(70)`,\n`pct_change(5) > 0.05`, `latest > 70 and prev <= 70`, etc. `channels` is a\nlist of names ([\"telegram\",\"slack\"]) or a {name: override} object. All\nchannel credentials live in root .env under ALERTS_*." }, { "name": "alerts_list_alert_rules", "description": "Return all alert rules." }, { "name": "alerts_get_alert_rule", "description": "Return one rule by name." }, { "name": "alerts_update_alert_rule", "description": "Update a rule's fields. Only provided fields change. The name cannot be renamed." }, { "name": "alerts_delete_alert_rule", "description": "Delete a rule. Its alert_events rows are removed via FK CASCADE." }, { "name": "alerts_list_channels", "description": "Return {name, configured, missing_keys} for each of the 7 channels\n(telegram, discord, slack, twitter, dingtalk, feishu, wecowork).\n\nNever returns credential values — only the names of missing keys." }, { "name": "alerts_run_alert_rule", "description": "Ad-hoc: evaluate one rule now and dispatch if it fires.\n\nRespects fire_mode + cooldown like the cron path. Use to test a rule on demand." }, { "name": "alerts_list_events", "description": "Recent alert_events (optionally for one rule), newest first." }, { "name": "cron_create_schedule", "description": "Create a new schedule with a cron expression and task." }, { "name": "cron_list_schedules", "description": "List all schedules." }, { "name": "cron_get_schedule", "description": "Get details for a single schedule." }, { "name": "cron_delete_schedule", "description": "Delete a schedule permanently." }, { "name": "cron_pause_schedule", "description": "Pause a schedule (disable without deleting)." }, { "name": "cron_resume_schedule", "description": "Resume a paused schedule." }, { "name": "cron_run_now", "description": "Execute a scheduled task immediately (one-shot)." }, { "name": "cron_list_executions", "description": "List execution history, optionally filtered by schedule_id." }, { "name": "cron_list_tasks", "description": "List available task types in the registry." }, { "name": "cron_create_task", "description": "Register a new task in the database. The command is a shell command or script path.\n\nExample: create_task(name=\"fetch_mofcom\", command=\"python3 scripts/fetch_mofcom_news.py --json\")" }, { "name": "cron_delete_task", "description": "Delete a task from the database by name." }, { "name": "cron_list_db_tasks", "description": "List all tasks stored in the database (user-defined, no code changes needed)." }, { "name": "cron_update_task", "description": "Update an existing DB task's command, description, or timeout. Only provided fields are updated." }, { "name": "composite_list", "description": "List all defined composites (name + description)." }, { "name": "composite_create", "description": "Create a new composite. Raises if the name already exists." }, { "name": "composite_list_upstreams", "description": "List the upstreams defined for a composite." }, { "name": "composite_add_upstream", "description": "Add an upstream MCP server to a composite.\n\nFor stdio: provide command, args, cwd (env optional).\nFor http: provide url." }, { "name": "composite_remove_upstream", "description": "Remove an upstream (and its selected tools) from a composite." }, { "name": "composite_list_available_tools", "description": "List the tools an upstream currently exposes (live), optionally filtered." }, { "name": "composite_add_tool", "description": "Select an upstream tool to be exposed (proxied) by the composite.\n\nThe tool is served as _ on next start.\n(alias is accepted but unused in v1.)" }, { "name": "composite_remove_tool", "description": "Remove a selected tool from a composite." }, { "name": "composite_list_tools", "description": "List the tools currently selected for a composite." }, { "name": "composite_add_chained_tool", "description": "Define a chained tool: a linear pipeline of upstream tool calls.\n\nEach step is {\"upstream\": , \"tool\": , \"input\": {param: value}}.\nInput values may be literals or references: \"$prev.\" (prior step's\nresult) or \"$step[N].\" (any prior step's result). Linear only — no\nbranching/conditionals." }, { "name": "composite_remove_chained_tool", "description": "Remove a chained tool from a composite." }, { "name": "composite_list_chained_tools", "description": "List the chained tools defined for a composite." }, { "name": "composite_create_manifest", "description": "Create a composite manifest: composite + upstreams + tools + workflows + prompt in one call." }, { "name": "composite_update_manifest", "description": "Partially update a composite manifest. Only provided fields change.\n\nupstreams/tools (when given) REPLACE the existing sets wholesale." }, { "name": "composite_delete_manifest", "description": "Delete a composite manifest (cascades upstreams/tools/chains)." }, { "name": "composite_list_manifests", "description": "List all composites as manifests (name, upstreams, tools, workflows, prompt)." }, { "name": "daas_list_sources", "description": "List all configured DAAS data sources with function counts.\n\nReturns sources like akshare, worldbank, ckan, cnstats\nwith install status and available function counts." }, { "name": "daas_search_functions", "description": "Search DAAS functions by name, category, or description across all sources." }, { "name": "daas_get_function_detail", "description": "Get full details for a DAAS function — parameters, output columns, description." }, { "name": "daas_list_categories", "description": "List all DAAS function categories with counts, optionally filtered by source." }, { "name": "daas_fetch_data", "description": "Execute a DAAS data function and return results as JSON.\n\nRoutes to the correct source adapter based on function name prefix." }, { "name": "daas_create_datasource", "description": "Create a new managed datasource." }, { "name": "daas_update_datasource", "description": "Update mutable fields of a datasource, identified by name or id.\n\nPass clear_category=true to unset the category (move to root level).\nPass score= to set the default score; pass clear_score=true to reset\nthe score back to NULL (unset). Only supplied fields are changed." }, { "name": "daas_delete_datasource", "description": "Delete a datasource and cascade-delete its forms, sections, and\ncollection-item references. Identify by name or id." }, { "name": "daas_create_category", "description": "Create a category, optionally under a parent (building the tree)." }, { "name": "daas_move_category", "description": "Re-parent a category. Rejects moves that would create a cycle\n(into own descendant) or self-parenting. Pass parent_id=null to make\nit a root." }, { "name": "daas_delete_category", "description": "Delete a category. Rejected if it has child categories. Datasources\nassigned to it are orphaned to root level (category_id set to null)." }, { "name": "daas_get_category_tree", "description": "Return the category tree (or a subtree from root_id) as a nested\nstructure, each node annotated with datasource_count." }, { "name": "daas_add_form", "description": "Add a form (e.g. '10-K', '8-K') to a datasource." }, { "name": "daas_add_section", "description": "Add a section to a form, carrying an extraction instruction." }, { "name": "daas_list_forms", "description": "List all forms of a datasource, each with nested sections\n(including instruction)." }, { "name": "daas_create_collection", "description": "Create a named datasource collection." }, { "name": "daas_add_to_collection", "description": "Add a datasource (or a specific datasource-section) to a collection.\n\nOmit section_name to add the whole datasource; supply section_name to\nadd only that section. Optional score sets a per-collection score override\non the new item (NULL = inherit the datasource's default score)." }, { "name": "daas_list_collection_items", "description": "List a collection's items, each resolved to source name and (if set)\nsection name + instruction." }, { "name": "daas_remove_from_collection", "description": "Remove an item from a collection by (collection, source, optional\nsection). Omit section_name to remove the whole-datasource item." }, { "name": "daas_list_collections", "description": "List every datasource collection with its item count." }, { "name": "daas_rename_collection", "description": "Rename a collection. The new name must be unique. Items are preserved\n(they reference by collection_id)." }, { "name": "daas_update_collection", "description": "Partially update a collection's name and/or description.\n\nAt least one of `new_name` / `description` must be provided; omitted\nfields are left unchanged. When `new_name` differs from the current\nname it must be unique. Raises if the collection is not found." }, { "name": "daas_delete_collection", "description": "Delete a collection. Cascades to its `datasource_collection_items`\nrows (datasources themselves are untouched)." }, { "name": "daas_reorder_collection_items", "description": "Rewrite the sort_order of items in `collection_name` to match the\ngiven list. `ordered_item_ids` MUST include every existing item id in\nthe collection exactly once — partial reorders are rejected." }, { "name": "daas_set_collection_item_score", "description": "Set or clear the per-collection score override on an existing collection\nitem, identified by (collection_name, source_name, optional section_name).\n\nPass score= to set the override; pass score=null (omit) to clear it\nso the item falls back to the datasource's default score. Returns the\nupdated item with its resolved effective score, the raw item_score override,\nand the source_default_score." }, { "name": "daas_search_datasources", "description": "Search/filter datasources across levels — category (with optional\nsubtree), source, form, and section — every filter optional, plus a\nfree-text query across source label/description, form label, and\nsection name/instruction." }, { "name": "daas_search_entities", "description": "Search entities (stocks + countries) by name, ticker, code, or alias." }, { "name": "daas_get_entity", "description": "Get full detail for one entity, including aliases and metadata.\n\nPer D5/3.5 the entity master is fd-open-data-mcp. Resolve by natural\nkey (entity_type, code) when given (gateway-first, local fallback);\notherwise fall back to entity_id (local)." }, { "name": "daas_list_entities", "description": "List entities filtered by type / exchange / country, paginated." }, { "name": "daas_get_entity_coverage", "description": "For each datasource linked to the entity, return the identifier to\nuse, the available sections (routing instructions = how to get the\ndata, with an identifier-prefilled variant), and the column count/list\naggregated from daas_function_columns. Sources without registered\nfunctions get a column_hint naming the sibling MCP + tool.\n\nPer D5/3.5 entity resolution proxies through fd-open-data-mcp. Accept\na natural key (entity_type, code) as an alternative to entity_id.\n\nThis answers: \"I have company X — which datasources cover it, how many\ncolumns can I get, and how do I fetch it?\"" }, { "name": "daas_link_entity_datasource", "description": "Link an entity to a daas datasource, recording the identifier to use\ninside that datasource (e.g. for AAPL → yfinance: 'AAPL'). Upserts on\n(entity_id, source_name)." }, { "name": "daas_unlink_entity_datasource", "description": "Delete the link between an entity and a datasource." }, { "name": "daas_create_entity_collection", "description": "Create a named entity collection (watchlist / portfolio)." }, { "name": "daas_list_entity_collections", "description": "List every entity collection with its item count." }, { "name": "daas_get_entity_collection", "description": "Get one collection with its current member entities (ordered)." }, { "name": "daas_update_entity_collection", "description": "Partially update an entity collection's name and/or description and/or\nrule. At least one field must be provided. `clear_rule=True` resets the\nrule to NULL (manual collection)." }, { "name": "daas_delete_entity_collection", "description": "Delete a collection and cascade to its items + changes." }, { "name": "daas_add_entity_to_collection", "description": "Add an entity to a collection. Resolves the entity by `entity_id`, or\nby `(entity_type, code)`. Records an `add_in` event in\n`entity_collection_changes` (source='manual'). No-op (no event recorded)\nif the entity is already a member." }, { "name": "daas_remove_entity_from_collection", "description": "Remove an entity from a collection. Records a `remove_out` event in\n`entity_collection_changes` (source='manual'). No-op (no event recorded)\nif the entity is not a member." }, { "name": "daas_list_entity_collection_items", "description": "List the current members of a collection, ordered by sort_order, each\nenriched with full entity detail (code, name, ticker, exchange, ...)." }, { "name": "daas_reorder_entity_collection_items", "description": "Rewrite member sort_order to match the given ordered list of\n`entity_collection_items.id`. Must contain exactly the current item ids." }, { "name": "daas_list_entity_collection_changes", "description": "Query the add-in / remove-out audit log (`entity_collection_changes`),\nnewest first, each enriched with collection name + entity code/name." }, { "name": "daas_sync_entity_collection", "description": "Re-derive the member set for a rule-based collection by applying its\n`rule_json` to the `entities` table, diff vs current members, apply\nadd_in for new matches and remove_out for non-matches (source='cron'),\nand record every transition. Returns `{added, removed, unchanged}`.\nA manual collection (rule_json=NULL) is a no-op." }, { "name": "daas_create_indicator_collection", "description": "Create a named indicator collection (a reusable bundle of indicators)." }, { "name": "daas_list_indicator_collections", "description": "List every indicator collection with its item count." }, { "name": "daas_get_indicator_collection", "description": "Get one collection with its current items (ordered, each with the\nresolved effective score + raw item/indicator/datasource scores)." }, { "name": "daas_update_indicator_collection", "description": "Partially update an indicator collection's name, description, and/or\nrule. At least one field must be provided. `clear_rule=True` resets\n`rule_id` to NULL (manual collection)." }, { "name": "daas_delete_indicator_collection", "description": "Delete a collection and cascade to its items + changes." }, { "name": "daas_add_indicator_to_collection", "description": "Add an indicator to a collection. Resolves the indicator by `name`." }, { "name": "daas_remove_indicator_from_collection", "description": "Remove an indicator from a collection. Records a `remove_out` event in\nthe audit log. No-op (`not_member`) if the indicator is not in the collection." }, { "name": "daas_list_indicator_collection_items", "description": "List the items of a collection, ordered by sort_order, each with the\nresolved effective score + raw item/indicator/datasource scores." }, { "name": "daas_reorder_indicator_collection_items", "description": "Rewrite item sort_order to match the given ordered list of\n`indicator_collection_items.id`. Must contain exactly the current item\nids of this collection — partial reorders are rejected." }, { "name": "daas_set_indicator_collection_item_score", "description": "Set or clear the per-collection score override on an existing item,\nidentified by (collection_name, indicator_name).\n\n`score` = a float sets the override; `score = null` clears it (the item\nthen inherits the indicator's default `indicator_rules.score`, which\nitself inherits the datasource default when NULL). Returns the updated\nitem dict with the resolved effective score." }, { "name": "daas_list_indicator_collection_changes", "description": "Query the add-in / remove-out audit log (`indicator_collection_changes`),\nnewest first. Each row is enriched with the collection name." }, { "name": "daas_sync_indicator_collection", "description": "Re-derive the member set for a rule-based indicator collection by\nevaluating its rule (target='indicator_names') via the RuleEngine, diff vs\ncurrent members, apply add_in/remove_out (source='cron'), and record every\ntransition. Returns `{added, removed, unchanged}`. A manual collection\n(rule_id NULL) is a no-op." }, { "name": "daas_create_pipeline_collection", "description": "Create a named pipeline collection." }, { "name": "daas_list_pipeline_collections", "description": "Every pipeline collection with its item count." }, { "name": "daas_get_pipeline_collection", "description": "One pipeline collection with all its items." }, { "name": "daas_delete_pipeline_collection", "description": "Unwire cron for every item, then delete the collection (FK CASCADE removes items)." }, { "name": "daas_list_pipeline_items", "description": "List items, optionally filtered to one collection." }, { "name": "daas_add_pipeline_item", "description": "Validate + store an item. If enabled and backfill, run history backfill\nthen register the cron-mcp schedule. Returns {item, backfill, cron}." }, { "name": "daas_remove_pipeline_item", "description": "Unwire cron then delete the item row. scraw_
is left intact." }, { "name": "daas_enable_pipeline_item", "description": "Backfill + register cron for a disabled item; set enabled=1." }, { "name": "daas_disable_pipeline_item", "description": "Delete the cron schedule (keep the task row); set enabled=0." }, { "name": "daas_update_pipeline_item", "description": "Update mutable fields. cron/timezone change re-syncs the schedule.\nenabled toggle runs enable/disable. arguments_json change does NOT backfill." }, { "name": "daas_sync_pipeline_cron", "description": "Re-apply cron wiring for all enabled items; remove schedules for disabled ones." }, { "name": "daas_list_models", "description": "Return configured models with a `vision` flag each. API keys are never returned." }, { "name": "daas_list_source_tables", "description": "Return scraped source-data tables (name LIKE 'scraw_%') with row counts + columns." }, { "name": "daas_create_rule", "description": "Create a rule." }, { "name": "daas_list_rules", "description": "List all rules, optionally filtered by rule_type." }, { "name": "daas_get_rule", "description": "Return one rule by name." }, { "name": "daas_update_rule", "description": "Partially update a rule. At least one field must be provided. The rule\nname cannot be renamed." }, { "name": "daas_delete_rule", "description": "Delete a rule. Referencing collections' rule_id is set to NULL; the\nrule's process_results rows are removed." }, { "name": "daas_run_rule", "description": "Evaluate + persist. For `target='rows'` (llm): incremental extraction\ninto process_results + advance last_rowid. For member targets: returns the\nderived set (membership is applied by daas_sync_*_collection, not here)." }, { "name": "daas_test_rule", "description": "Dry-run: evaluate the rule WITHOUT persisting. For member targets\n(entity_ids/indicator_names) returns the derived set. For `target='rows'`\n(llm), extracts from a single source row (no process_results write) as a\nsample." }, { "name": "daas_extract_text", "description": "Extract structured records from (possibly long) text.\n\nLong text is chunked (no truncation): per-chunk extraction then a merge pass." }, { "name": "daas_extract_image", "description": "Extract structured records from an image via a vision-capable model.\n\n`image` accepts a local file path, an http(s):// URL, or raw base64." }, { "name": "daas_extract_file", "description": "Read a local .txt/.md/.pdf and extract structured records." }, { "name": "daas_list_indicator_ops", "description": "Return the fixed math-op catalog with each op's required params." }, { "name": "daas_create_indicator", "description": "Create an indicator rule bound to a datasource + source table." }, { "name": "daas_list_indicators", "description": "Return all indicator rules (each with `score` + `effective_default_score`)." }, { "name": "daas_get_indicator", "description": "Return one indicator rule by name (with `score` + `effective_default_score`)." }, { "name": "daas_update_indicator", "description": "Update an indicator rule's fields. Only provided fields change. The\nrule name cannot be renamed. Pass `params` to replace the op params.\n\n`score`: a float sets the indicator's default score. `clear_score=True`\nclears it back to NULL (inherit the datasource's `sources.score`).\nMirrors `update_datasource(score=…, clear_score=True)`." }, { "name": "daas_set_indicator_score", "description": "Set the indicator's default `score` (float), or clear it (None → inherit\nthe datasource's `sources.score`). Returns the updated indicator dict with\n`effective_default_score`." }, { "name": "daas_delete_indicator", "description": "Delete an indicator rule. observations rows it produced are NOT removed\n(soft reference) — they remain identifiable via their `metadata.rule_name`." }, { "name": "daas_run_indicator", "description": "Full-recompute the indicator over its source table → observations.\n\nReads the whole source table, computes the op, and upserts every\n(date, value) into the daas `observations` table keyed on\n(source=datasource, function_name, indicator=indicator_name, date).\nIdempotent on re-run via the observations unique constraint." }, { "name": "daas_calculate_indicator", "description": "Ad-hoc: compute an indicator over a source table without persisting.\n\nValidates table/columns/op/params, reads the series, computes, and returns\n{indicator, dates, values, count}. Writes nothing (no rule, no observations)." }, { "name": "dashboard_list_databases", "description": "List all known SQLite databases in the MCP directory with table counts." }, { "name": "dashboard_query_table", "description": "Run a read-only query against a database table. Returns rows as JSON." }, { "name": "dashboard_list_datasources", "description": "List managed datasources from daas.db (schema managed by mcp/models/models.py)." }, { "name": "dashboard_get_executions", "description": "Get cron execution history from a database." }, { "name": "dashboard_get_stats", "description": "Get aggregated statistics across all MCP databases." }, { "name": "dashboard_register", "description": "Register or update (upsert by slug) a standalone HTML dashboard in the\n`dashboards` table. After writing the row, regenerates index.html + daas.md\nfrom the DB. `source_tables`/`entity_coverage`/`time_range`/`chart_config`\naccept a JSON string or a list/dict. Returns the stored row (with `action`\n= inserted|updated) or `{\"error\": ...}`." }, { "name": "dashboard_list", "description": "List every registered dashboard (name + slug + intro + file_url), oldest\nfirst. Returns a JSON array." }, { "name": "dashboard_get", "description": "Get one dashboard's full metadata by slug: name, intro, source_tables,\nentity_coverage, time_range, refresh_cadence, chart_config, file_path,\nfile_url. Returns `{\"error\": \"dashboard '' not found\"}` if absent." }, { "name": "dashboard_search", "description": "Case-insensitive keyword search over each dashboard's name + intro +\nrefresh_cadence + source_tables. Returns matching `{slug, name, intro}`\nentries, or an empty array if none match." }, { "name": "dashboard_update", "description": "Patch one or more fields of an existing dashboard (by slug). Only the\nfields you pass are changed; the rest are left as-is. Regenerates index.html\n+ daas.md after the update. Returns the updated row or `{\"error\": ...}`." }, { "name": "dashboard_delete", "description": "Delete a dashboard row by slug and regenerate index.html + daas.md. The\nHTML file itself is NOT removed (only the registry row). Returns\n`{\"deleted\": slug}` or `{\"error\": ...}`." }, { "name": "gateway_list_data_mcps", "description": "List the data-fetch MCP upstreams gateway-mcp can route to.\n\nEach entry is a row from the `gateway_upstreams` table (name, transport,\nenabled, description). Disabled upstreams are hidden unless\n`include_disabled=True`." }, { "name": "gateway_list_data_mcp_tools", "description": "List the tools exposed by a data-fetch MCP upstream (live).\n\nReuses the persistent client from the connection pool — the client stays\nalive across calls and is only recreated when the upstream config changes.\nReturns each tool's name + description (+ parameters when available)." }, { "name": "gateway_call_data_mcp", "description": "Call a tool on a data-fetch MCP upstream and return its result.\n\nConnects to the named upstream via the persistent client pool, invokes\nthe named tool with the JSON-deserialized `arguments`, and returns the\nupstream's result. The client stays alive across calls and is only\nrecreated when the upstream config changes." }, { "name": "gateway_add_data_mcp", "description": "Add or update a data-fetch MCP upstream (idempotent upsert by name)." }, { "name": "gateway_remove_data_mcp", "description": "Remove a data-fetch MCP upstream from the registry." }, { "name": "gateway_get_data_mcp", "description": "Get one data-fetch MCP upstream's full launch config." }, { "name": "gateway_gateway_health", "description": "Health-probe each enabled gateway upstream and auto-flip transport.\n\nFor each enabled upstream in ``gateway_upstreams``:\n\n- **http** transport: ping the row's ``url``. On failure AND the row\n carries a stdio ``command``, flip ``transport`` to ``stdio`` (degraded\n mode — mirrors the client-pool fallback so the next ``get_client()``\n launches the subprocess). On success, no change.\n- **stdio** transport with a ``url`` set (i.e. it was flipped from http by\n a prior fallback): ping the ``url``. On success flip ``transport`` back\n to ``http`` (recovered). On failure, leave as stdio.\n- **stdio**-only (no ``url``): skipped (nothing to probe).\n\nReturns a per-upstream status report. Never raises — network errors are\ncaptured as degraded states so the probe is safe to call from selfcheck." }, { "name": "research_create", "description": "Create a persisted research bundle. Named collections are attached by\nname; with ``create_missing=True`` missing entity/indicator/pipeline\ncollections are created empty. A ``dashboard_slug`` must already exist\n(the tool does not build dashboard HTML). ``component_refs`` is a JSON\nobject string: ``{\"rules\": [...], \"scraw_tables\": [...], \"indicators\": [...]}``." }, { "name": "research_get", "description": "Return a research + resolved bundle (live counts, dashboard url,\npipeline status, dangling-component warnings)." }, { "name": "research_list", "description": "List every research (optionally filtered by status) with attached\ncomponent names." }, { "name": "research_update", "description": "Partially update a research. ``None`` means leave a field unchanged;\nto detach a named component use ``remove_component``. Any newly-set\ncomponent name is validated to exist. ``component_refs`` (JSON string)\nreplaces the auxiliary refs." }, { "name": "research_delete", "description": "Delete a research. Removes the on-disk report file. With\n``remove_pipeline=True`` (default) deletes the owned pipeline collection\n+ orphaned cron schedules/tasks. With ``remove_dashboard=True`` (default\nFalse) deletes the referenced dashboard via the sibling tool. Shared\nentity/indicator collections, rules, scraw tables, and observations are\nNEVER deleted." }, { "name": "research_generate_report", "description": "Assemble a markdown report from the bundle's live components, persist it\ninto ``report_md`` and write it to ``researches/.md`` (dir from\n``RESEARCH_DIR``, default repo-root ``researches/``). Regeneration\noverwrites both. Returns ``{name, report_path, char_count}``." }, { "name": "research_refresh", "description": "Re-run the research's data pipeline: recompute each indicator in the\nindicator collection (via ``daas_run_indicator``), sync rule-based\ncollections (via ``daas_sync_entity_collection``/``daas_sync_indicator_collection``),\nand report each pipeline item's current status. Returns a per-component\nstatus report; per-item errors are reported, not raised." }, { "name": "research_add_component", "description": "Attach an existing component to a research by name. ``component_type`` ∈\nentity_collection | indicator_collection | dashboard | pipeline | rule |\nscraw_table | indicator. The underlying object is not created - it must\nalready exist (for the named kinds)." }, { "name": "research_remove_component", "description": "Detach a component from a research. The underlying object is NOT\ndeleted. For named kinds, clears the research column; for auxiliary refs,\nremoves the name from ``component_refs``." }, { "name": "workflow_register", "description": "Register a new workflow manifest (YAML). Version bumps on name collision\nunless the manifest declares its own ``version``." }, { "name": "workflow_get", "description": "Get a workflow by name (latest version by default)." }, { "name": "workflow_list", "description": "List every workflow (name, version, description, enabled, step_count)." }, { "name": "workflow_update", "description": "Patch a workflow. Only provided fields change. ``version`` defaults to\nthe latest version of ``name``." }, { "name": "workflow_delete", "description": "Delete a workflow. Omit ``version`` to delete every version of ``name``.\nSteps/runs cascade via FK." }, { "name": "workflow_run", "description": "Run a workflow manifest. ``params_json`` merges over the manifest's\ndeclared ``params``. Returns run id + status; a paused run carries a\n``resume_token`` for ``resume``." }, { "name": "workflow_inspect", "description": "Validate a stored manifest and return its parsed step plan (no execution)." }, { "name": "workflow_resume", "description": "Resume a paused run. ``approved=False`` marks it failed." }, { "name": "customer_dataset_ingest_begin", "description": "Begin a full-snapshot ingest for a customer dataset (registers/refreshes\nthe dataset metadata and creates the staging table). Supersedes any\nin-flight ingest for the same key." }, { "name": "customer_dataset_ingest_chunk", "description": "Append one chunk of rows to an open ingest. `start_index` must equal the\nnumber of rows already staged (guards against replays)." }, { "name": "customer_dataset_ingest_commit", "description": "Commit an open ingest per its begin-time mode: `replace` atomically\nswaps the staged snapshot into the base table; `merge` upserts the staged\nrows into the base by the dataset's primary key (rows absent from staging\nare kept). Either way the effective table is rebuilt and all active\ncorrections replayed (dangling anchors are marked, never dropped). Returns\nrows/size/corrections stats." }, { "name": "customer_dataset_ingest_abort", "description": "Abort an open ingest and drop its staging table (base/effective stay)." }, { "name": "customer_dataset_overview", "description": "Dataset metadata: columns, pk_columns, rows, size_bytes, last_commit_at.\nWhen FD_CELL_STORAGE_QUOTA_MB is armed on the PostgreSQL backend, also\nstorage_quota_bytes / storage_used_bytes (pg_database_size)." }, { "name": "customer_dataset_preview", "description": "Paged row preview with `__row_id` anchors. `include_corrected=false`\nreads the raw snapshot (original values stay queryable)." }, { "name": "customer_dataset_query", "description": "Structured query over a customer dataset: whitelisted filter operators\n(eq/ne/gt/gte/lt/lte/in/notin/like/isnull/notnull, AND-composed), multi-column\nsort, limit/offset paging, exact total and an honest truncated flag; rows\ncarry `__row_id` anchors. `count_only=True` returns just the count;\n`distinct_column` returns that column's distinct values (capped at 1,000);\n`raw=True` reads the base snapshot without edits. Responses are capped at\n1,000 rows and queries abort past 5s." }, { "name": "customer_dataset_aggregate", "description": "GROUP BY aggregation over a customer dataset: multi-column grouping ×\ncount/sum/avg/min/max measures, filter-compatible, group rows capped at\n1,000 with exact group_count. sum/avg reject TEXT columns (honest type\nguard); min/max allow any column. `raw=True` reads the base snapshot." }, { "name": "customer_dataset_correction_add", "description": "Add an overlay correction (update/delete/insert); it applies to the\neffective table immediately and is replayed after every future commit.\nEvery edit record carries its write channel (console | data-key | agent)\nfor audit attribution." }, { "name": "customer_dataset_correction_list", "description": "Correction history (audit trail) for a dataset." }, { "name": "customer_dataset_correction_dangling", "description": "Corrections whose row anchor no longer matches the latest snapshot." }, { "name": "customer_dataset_correction_revert", "description": "Revert one correction (mark reverted; rebuild the effective table from\nthe base snapshot plus the remaining active corrections)." }, { "name": "customer_dataset_write_batch", "description": "Business-write surface (wire-customer-data-engine §2): a list of edit\noperations applied in ONE transaction (≤1,000 ops). Any rejected op rolls\nthe whole batch back — zero partial writes — and the error carries the\n1-based index of the failing entry. Anchors are strict here: an\nupdate/delete that matches no row rejects the batch (business writes must\nhit their anchors; dangling is a connector-world outcome, not an app one)." }, { "name": "customer_dataset_create_declared", "description": "Schema-declared dataset creation: an EMPTY dataset from a column/type\ndeclaration (optionally with primary-key columns that become the row\nanchor), indistinguishable from an uploaded/connector-created dataset.\nThe app-first creation channel for business systems." }, { "name": "customer_dataset_add_column", "description": "Add ONE nullable column to a dataset (effective + base tables and the\nmetadata). Additive-only schema evolution: destructive changes are not\noffered. Rejected while an ingest is open." }, { "name": "customer_dataset_derive", "description": "Materialize a structured-query result over `parent_key` into a NEW\nchild dataset, lineage recorded (parent + spec). The child inherits the\nparent's columns and pk declaration. Re-derivation of an existing child\ngoes through `derive_refresh`." }, { "name": "customer_dataset_join_indicators", "description": "Derive a child dataset = customer rows × platform indicator series.\nSeries are fetched through the wire entrance with the cell's credential\n(metered to this customer, same track as artifact materialization).\nMatching: exact date → month period → year period (frequency inferred\nfrom the series' own points); misses join as NULL — never neighbor\nvalues. `region_column` (optional) resolves per-row entities by name;\nwithout it every row uses `default_entity`. At most 10 indicator codes\nper derivation and 64 distinct regions." }, { "name": "customer_dataset_link", "description": "Materialize a semantic connection between two customer datasets into a\nNEW child dataset (wire-entity-link): LEFT JOIN semantics — every left row\nis kept; rows whose key has no right counterpart carry NULL in the right\ncolumns (never a neighbor's value). Right columns whose name collides with\na left column (or an already-chosen right output name) are renamed\n``right_`` (then ``right__2`` …); uncolliding right columns\nkeep their names and types. The child inherits the LEFT dataset's pk\ndeclaration and its lineage is recorded as kind=\"link\" (parent = left\ndataset; the right dataset and both join columns live in the spec), so\n``derive_refresh`` re-joins both sides' latest effective state. The join\nruns on the effective tables (corrections apply) using SQL ``IS`` key\nequality — NULL keys compare equal to NULL, keeping equality semantics\nconsistent. Row counts beyond the cap reject the whole operation (the\nchild keeps its previous state).\n\nm:n 桥展开（wire-entity-mn-materialize）：给齐 junction 三要素\n（junction_key / junction_left_column / junction_right_column）时改走\n桥行驱动 INNER 对表——一条桥行一行结果、无桥关联端点行不入结果、\n中间集列不进输出、孤儿桥以 ``orphan_dropped`` 计数如实上报；三要素\n必须同给同缺。缺省时 n:1/1:1 行为逐字节等价。" }, { "name": "customer_dataset_derive_refresh", "description": "Re-run a derived dataset's stored spec against its parent's current\neffective state. Join derivations re-fetch their indicator series first\n(metered reads again); link derivations re-join both sides from the\nrecorded spec; operator (filter/project/aggregate/union) and SQL\nderivations replay their archived spec verbatim." }, { "name": "customer_dataset_lineage", "description": "Lineage of a dataset: its own derivation (parent + spec + last run)\nand its live derived children." }, { "name": "customer_dataset_delete", "description": "Delete a dataset: drop its three tables, metadata, corrections and\nsoft-reference rows. Ingest audit rows are kept." }, { "name": "customer_dataset_filter", "description": "Materialize a filtered (optionally sorted/limited) subset of\n`parent_key` into a NEW child dataset. Same filter/sort/limit semantics\nas the query surface and as `derive` — the difference is lineage kind\n`filter`, so pipelines and the dataset graph label this step truthfully\ninstead of calling it a generic derivation." }, { "name": "customer_dataset_project", "description": "Materialize a column projection (with renaming) into a NEW child\ndataset. Output column order follows the declaration order. Primary-key\ncolumns that are projected follow their new names; keys left out of the\nprojection are dropped (row anchoring then falls back to the row hash)." }, { "name": "customer_dataset_aggregate_derive", "description": "Materialize a GROUP BY result into a NEW child dataset (rows = groups,\ncolumns = group keys + measures). This is the materializing counterpart of\nthe read-only `aggregate` tool: same validation and type guards, but the\noutput becomes a real dataset that downstream steps can build on.\nsum/avg reject TEXT columns (honest type guard); min/max allow any column." }, { "name": "customer_dataset_union", "description": "Append two datasets vertically into a NEW child dataset, aligning\ncolumns by name: column set = union of both sides (first-seen order),\nmissing columns become NULL. Same-name columns with incompatible types\n(TEXT vs numeric) are rejected honestly — never silently coerced;\nINTEGER + REAL widens to REAL. The output has no primary key (append\nsemantics)." }, { "name": "customer_dataset_window", "description": "Materialize a partitioned/ordered window computation over `parent_key`\ninto a NEW child dataset: source columns passed through unchanged plus one\ncolumn per `window` row (named by `as`). The row set is unchanged (windows\nnever add or drop rows); lineage kind `window` records the full spec so\n`derive_refresh` replays it.\n\n函数闭集（design D1）：row_number/rank/dense_rank 与 lag/lead 必须有\n`order_by`（否则 422 指名函数）；lag/lead 的 `offset` 缺省 1、必须\n1..10000；sum/avg/min/max 的 `column` 必填且 ⊆ 列快照，count 允许无\ncolumn（全行计数）。`as` 必须给出、唯一、且不与源列名冲突（422 指名，\n不静默改名）。`frame ∈ whole_partition|running`（running 必须有\norder_by；闭集，不开放任意边界语法）；frame 只对聚合类生效——行级函数\n（row_number/rank/dense_rank/lag/lead）的取值与帧无关。\n\n确定性（design D2）：内部 ORDER BY 末尾追加稳定行键 `__row_hash` 作\n最终 tie-break，并列数据重派生逐行一致；`__row_hash` 与一切保留前缀列\n绝不进输出列。" }, { "name": "customer_dataset_sql", "description": "Materialize a read-only SQL query into a NEW child dataset. The SQL\ntext, placeholder mapping and parameters are archived in lineage\n(append-only) so the step can be replayed and audited; replay re-executes\nthe archived statement verbatim.\n\nRead-only by construction: single statement starting with SELECT/WITH,\nno DDL/DML/PRAGMA/ATTACH, dangerous functions denied, a sqlite authorizer\nthat denies anything but reads of the declared datasets, and the\nconnection itself opened in read-only mode. Results over the dataset row\nlimit reject the whole batch." }, { "name": "customer_dataset_sql_preview", "description": "Dry-run a read-only SQL step: returns at most 100 preview rows plus\nthe result-set schema and creates/refreshes NOTHING. Guardrails are\nidentical to `sql_derive` (single read-only statement, declared\nplaceholders only, literal parameter binding, dangerous functions\ndenied, read-only connection, timeout and row caps)." }, { "name": "customer_dataset_provenance_wal_pending", "description": "Read un-forwarded WAL events for the platform-side provenance\nforwarder: rows with ``acked = 0`` and ``seq > after_seq``, ascending by\n``seq`` (cursor style — hand the returned ``last_seq`` to\n``provenance_wal_ack``, then keep polling with it as ``after_seq``).\n``limit`` is clamped to [1, 1000]. Events stay readable until acked, so a\nforwarder crash between read and ack loses nothing. On a database that\npredates the WAL (table not yet created) returns an empty page." }, { "name": "customer_dataset_provenance_wal_ack", "description": "Ack WAL events up to and including ``upto_seq`` (forwarder confirms\nthey are durably in the platform ledger). Idempotent: re-acking returns\n``{\"acked\": 0}``. Flipping the ``acked`` bit is the only mutation the WAL\never allows — tamper-proofing lives at the platform ledger, not here." }, { "name": "customer_dataset_check", "description": "Batch content-level checks over a customer dataset\n(openspec: wire-quality-rules-v2). Each check: {kind, ...}:\n- regex: {kind: \"regex\", column, pattern} — pattern length ≤128, no\n nested quantifiers, executed with a deterministic timeout (regex lib);\n- col_compare: {kind: \"col_compare\", col_a, op, col_b} — ops:\n eq/ne/gt/gte/lt/lte;\n- range: {kind: \"range\", column, min?/max?} — closed interval, at least\n one bound.\n\nAt most 5 checks per call. Returns per-check Great-Expectations-style\noutcome {status: pass|fail|error, element_count, unexpected_count,\npass_rate}. NULL values never count as violations (null coverage is a\nseparate concern); an empty dataset passes vacuously. A check whose own\nexecution fails (e.g. regex timeout) reports `error`, never `fail`.\nFailure samples (offending values / row indexes) are aggregated to\ncounts inside the engine and discarded — the response contains no data\nvalues." }, { "name": "customer_dataset_quality", "description": "Quality snapshot aggregates for one dataset (openspec:\nwire-quality-rules-v2; payload shape aligned field-by-field with the\nplatform-side sanitizer). Returns {row_count, computed_at, pk_declared,\npk_uniqueness: {columns, unique_ratio} | null, null_rates: [{column,\nnull_rate, sampled_rows}], sample_limit_hit}. Only aggregate counts and\nratios — never data values." } ], "toolCount": 193, "license": null, "provider": null, "endpoint": "https://mcp.finddatatech.cloud/fd-wire-cell-pgauto001/mcp" } ] }