菜单
Access documents attached to a context (organization, folder, initiative, chat session) with list, summary, content, and analyze actions.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when bringing the harness method to a NEW or non-sentropic repo — maps the repo's conventions into a harness profile so the same scope/branch/verify kernel applies anywhere.
Use before any creative/design work (a feature, a package, an evolution) — explores intent and design via the spec-ladder with multi-peer adversarial review BEFORE implementation.
Use on any bug, test failure, or unexpected behaviour BEFORE proposing a fix — runs the generic evidence-first root-cause loop.
Use when you have a spec/decisions and need an executable plan — writes a lot-based BRANCH.md from the template and completes with track for realization state.
Use when completing a feature, a design step, or before merging — runs a ≥2-peer consensus review and reconciles the findings, instead of a single rubber-stamp pass.
Use when adding or evolving tests for a feature/bugfix — applies the sentropic test-pyramid, scoped-test loop, environment isolation, and AI-flaky policy.
| name | documents |
| description | Access documents attached to a context (organization, folder, initiative, chat session) with list, summary, content, and analyze actions. |
| version | 0.1.0 |
| category | content |
| tools | [{"name":"documents","description":"Accède aux documents attachés à un contexte (organization/folder/usecase/chat_session).\n\nLes documents sont fournis par l'utilisateurs et supposés être une source importante d'information.\n\nPermet de:\n - lister les documents + statuts,\n - lire un RÉSUMÉ COURT (get_summary) pour une information de surface,\n - lire le CONTENU (get_content) : soit le text complet (petit doc) soit un résumé (10k mots si le document est long) - pour une information détaillée\n - lancer une analyse ciblée (analyze) - pour une requête ciblée (recherche d'un contenu).\n","inputSchema":{"type":"object","properties":{"action":{"type":"string","enum":["list","get_summary","get_content","analyze"],"description":"Action à effectuer."},"contextType":{"type":"string","enum":["organization","folder","initiative","chat_session"],"description":"Type du contexte."},"contextId":{"type":"string","description":"ID du contexte."},"documentId":{"type":"string","description":"ID du document (requis pour get_summary/get_content)."},"maxChars":{"type":"number","description":"Optionnel: borne de caractères pour get_content (max 50000)."},"prompt":{"type":"string","description":"Requis pour analyze: prompt/instruction ciblée à exécuter par un sous-agent à partir du document (texte intégral si possible; sinon scan complet par extraits + consolidation)."},"maxWords":{"type":"number","description":"Optionnel: borne en mots pour analyze (max 10000, défaut 10000)."}},"required":["action","contextType","contextId"]}}] |
The documents skill exposes attachments stored under a workspace context
(organization, folder, initiative, or chat session). It supports four actions:
list documents and statuses, fetch a short summary, fetch bounded content, or
run a targeted sub-agent analysis over the document.
Handlers are intentionally not bound in this package commit. Runtime execution
still routes through the legacy API tool service until BR-19 Lot 5 rebinds
chat-service to SkillsToolRegistry.
