docs-governance
Skill320 starsChanged 33 days agoMITView on GitHub
作为面向长期 AI 协作项目的知识、决策与验证治理总入口,把优秀 Agent 的一次性工作沉淀为可继承、可验证、可持续演进的项目集体能力;根据用户意图把任务路由到活文档、领域上下文与 ADR、变更影响、接口契约、测试资产、模块回归或闭环设计能力,并在大型变更中组织正确顺序。用于用户只说“文档治理”“项目治理”“帮我整理项目知识”“改完怎么收尾”而未指定具体 Skill,或需要跨多项治理能力时。English triggers: docs governance router, project knowledge governance, documentation workflow, governance workflow.
8 other projects use a version of this.
Adopted by
Most recently taken up by
- hushh-labs/hushh-researchtheir own version, first seen Nov 1, 2026
- ghvinashvili/specagenttheir own version, first seen Nov 1, 2026
- CherryHQ/cherry-studiotheir own version, first seen Oct 25, 2026
- stainless-code/losslesstheir own version, first seen Oct 24, 2026
- Hexsis-LLC/skillstheir own version, first seen Oct 13, 2026
- qshanx/docs-governanceword for word, first seen Oct 13, 2026
Other versions8 projects use a version of this.
Word for word · 1
- qshanx/docs-governance★ 253100% same
Edited · 0
None found.
Wrote their own version · 7
- CherryHQ/cherry-studio★ 53ksame name
- clawdotnet/openclaw.net★ 533same name
- hushh-labs/hushh-research★ 40same name
- ghvinashvili/specagent★ 7same name
- stainless-code/lossless★ 4same name
- Hexsis-LLC/skills★ 1same name
- EraserandRain/skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository20 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- agent-entrypointsSkill
- agent-workflow-governanceSkill
- architecture-docsSkill
- context-and-decisionsSkill
- docs-auditSkill
- docs-maintenanceSkill
- docs-retroSkill
- docs-structureSkill
- domain-modelingSkill
- living-docs-governanceSkill
- module-governanceSkill
- product-evolutionSkill
- writing-for-agentsSkill
- change-impactSkill
- contract-firstSkill
- loop-design-checkSkill
- module-regressionSkill
- test-collaborationSkill
The file
Files
文档治理总路由
定位
把优秀 Agent 的一次性工作,沉淀为项目可继承、可验证、可持续演进的集体能力。
本插件不是让 Agent 多写文档,而是治理长期 AI 协作中的项目知识、关键决策与验证证据:让后续 Agent 能恢复上下文、沿用已确认约束、复核结果,并把新的有效做法继续沉淀回项目。
先判断意图,再读取并执行对应 Skill。不要在本路由复制各 Skill 的方法论。
单项路由
| 用户意图 | 路由到 |
| 新建核心模块、补齐模块文档或维护模块约定,无须点名 Skill | module-governance;讨论只读,普通 Bug/样式不启动建档 |
| 审计核心模块知识入口覆盖 | docs-audit,仅报告;边界核对使用 architecture-docs |
| 检查治理规则应放在入口、Skill、附件、脚本还是 Hook,识别散落的可复用工作流 | agent-workflow-governance,默认只读交付证据与建议 |
| 采集项目健康证据、核对验证入口与 CI 衔接 | living-docs-governance 的“项目健康证据采集” |
| 编写或优化 Skill、Agent 入口及 AI 读取的文档 | writing-for-agents;事实核对仍走审计流程 |
| 规划、整理或只读审查文档目录、归属与读取入口 | docs-structure |
| 生成、精简或审查 AGENTS.md / CLAUDE.md,明确项目规范和产品读取入口 | agent-entrypoints |
| 产品文档管理员、来源登记、需求池/编号/版本、十阶段管理与产品审查 | product-evolution;Claude Code 可由 product-docs-manager 执行 |
| 新项目、讨论后项目或已有项目首次配置产品文档包与 Agent 入口 | living-docs-governance 的“统一 setup”模式;Claude Code 使用 /governance-setup,/governance-init 是兼容别名 |
| 本次改动后的文档同步、整理、收尾或交接 | docs-maintenance |
| 从对话、文档变更与审计记录复盘文档管理经验 | docs-retro |
| LOG 管理或复盘 | living-docs-governance |
| 是否需要架构文档,生成、完善或审查模块分工、状态归属、依赖与核心流程 | architecture-docs |
领域术语、CONTEXT.md、架构或数据库决策、ADR | domain-modeling(Matt 原文与本地适配);context-and-decisions 仅兼容旧调用 |
| 修改前判断牵连面、迁移、回滚、实施后对照 | change-impact |
| 前后端或服务间接口 | contract-first |
| 测试资产盘点、重要覆盖缺口或契约证据 | test-collaboration |
| 修改后验证本模块及下游 | module-regression |
| 文档散乱、导航、时效、重复主源与当前/历史区分的只读审计 | docs-audit |
| 指定 spine/context/adr/artifacts/full 的兼容审计 | living-docs-governance 的审计模式 |
| 设计可判定目标和反馈回路 | loop-design-check |
模块变更流程
涉及跨模块行为、公共契约、共享数据或影响不明的变更前,当前 Agent 按需使用以下方法。普通小任务只改相关代码并运行与影响、风险相称的检查;文档事实未变就不更新,也不因收尾运行全量审计。MAP 只在路径、入口或定位关系变化时更新;LOG 只记值得追溯的重要事件;TESTS 只在测试入口、使用方法或重要覆盖范围变化时维护,不逐例登记或重复抄写运行结果。 小修范围已知时直接实施和验证,不逐个调用治理 Skill;只读请求保持只读。
- 改前看影响:记录任务起点及已有未提交修改,按
change-impact查真实调用方、数据与契约,确定受影响模块、候选文档和验证命令。需求不清先定位已确认 Spec/Issue;跨端接口按contract-first处理,难回退决策按domain-modeling处理。 - 实施并核对范围:完成获准修改,对照实际 diff 补查新增影响。任务开始时已有修改须单独区分,不能全归为本轮;出现新的需求或授权冲突时先澄清该部分。
- 验证本模块与下游:按
module-regression执行影响分析确定的回归;测试点与证据需要维护时调用test-collaboration。本次引入的失败由当前改动者修复并复验;旧失败、环境阻塞与未跑范围分别说明,不能把局部通过说成全部通过。 - 按需同步并交付:实际改变了文档承载的事实时,将任务范围、diff、影响清单和回归结果交给
docs-maintenance;需编写或优化面向 AI 的正文时,按writing-for-agents核对改写含义。没有文档事实变化则直接交付,不另建报告、不联动更新全部载体。复用同一基线与有效检查结果;新变化只补相关验证,既有门禁照常执行。最终简述变化、影响、验证及剩余缺口;回归受阻不能把功能写成验收通过。
接入项目入口
由 agent-entrypoints 在已授权配置的共享入口保留一句路标:
涉及跨模块行为、公共契约、共享数据或影响不明的变更前,读取 docs-governance 的“模块变更流程”;普通小任务直接修改、运行相关检查,文档事实未变不更新。
将 Skill 名替换为实际可读取路径;外部插件使用完整路径。setup 通过 agent-entrypoints 接入;已有项目只补这条路标,不复制流程正文。Skill 存在不等于自动运行:这是 Agent 执行约定,未安装 Hook/CI 就不声称强制触发。
边界
- 让
CLAUDE_MAP.md管项目知识位置,让ARCHITECTURE.md管当前 Module 结构;本 Skill 只管插件能力路由。 - 让 Spec/Issue 管业务成功标准,Issue Tracker 管任务状态和排期;不要复制进路由或数据库。
- 不因为用户说“治理”就一次性创建所有可选文档和目录。先发现现有事实载体,再按预警信号懒创建。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

