project-governance
为 AI 长期项目建立「项目记忆 + 文件索引 + 工作规则 + 版本记录」的治理系统,让 AI 在长期项目里「不忘事、不乱改、不重复犯错」,换会话后仍能接着做。Set up and maintain a project governance workspace for AI-assisted…
Skill by trae-community
Chalk Labs
Need help deploying project-governance?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/trae-community-project-governance
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 26, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Python package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/project-governance
What's inside
- Skill name
project-governance- Description
- 为 AI 长期项目建立「项目记忆 + 文件索引 + 工作规则 + 版本记录」的治理系统,让 AI 在长期项目里「不忘事、不乱改、不重复犯错」,换会话后仍能接着做。Set up and maintain a project governance workspace for AI-assisted long-running projects — project protocol (AGENTS.md), directory index (index.md), error log (LESSONS.md), session handoff, changelog, stable version index (VERSIONS.md), whitelist/blacklist parameter registries. Use when the user complains the project is messy, files are scattered or misplaced, the AI repeats mistakes or uses wrong versions, the user expects the AI to find files itself instead of asking for paths, when starting a new AI-assisted project, onboarding an AI agent into an existing project, or when a project lacks structured rules/versioning.
- License
- MIT-0
- Instructions
- 152 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Files (14)
- README.md
- SKILL.md
- governance.py · scripts/
- AGENTS.md · templates/
- ARCHITECTURE.md · templates/
- CHANGELOG.md · templates/
- LESSONS.md · templates/
- PROJECT.md · templates/
- VERSIONS.md · templates/
- blacklist.json · templates/
- index.md · templates/
- index_notes.json · templates/
- session_handoff.md · templates/
- whitelist.json · templates/
SKILL.md (first lines)
# Project Governance > 给 AI 长期项目建立一套「项目记忆 + 文件索引 + 工作规则 + 版本记录」的管理系统,让 AI 换会话、换模型后仍能正确接着项目做,而不是重新猜项目。 ## 目录 - [30 秒概览](#30-秒概览) - [何时使用 / 何时不用](#何时使用--何时不用) - [记忆与治理的边界](#记忆与治理的边界) - [使用步骤](#使用步骤) - [输入 / 输出](#输入--输出) - [失败处理](#失败处理) - [产出物](#产出物) - [包结构](#包结构) - [跨平台兼容](#跨平台兼容) - [限制](#限制) ## 30 秒概览 这个 Skill 给项目增加 8 个东西: 1. 项目规则 —— AI 应该怎么做 2. 文件地图 —— 文件在哪里 3. 项目状态 —— 现在做到哪里 4. 错误记录 —— 以前踩过什么坑 5. 版本索引 —— 哪个版本才是真的 6. 黑白名单 —— 什么能用、什么不能用 7. 会话交接 —— 上一个 AI 做到哪里 8. 变更记录 —— 为什么这么改 核心目标:**让 AI 换会话、换模型、甚至换 Agent 后,仍然能正确接着项目做,而不是重新猜项目。** ## 何时使用 / 何时不用 ### 何时使用 - 用户抱怨项目太乱、文件散落、AI 总是乱放文件("你怎么又乱放文件")。 - 用户期望 AI 自己找文件,而不是问路径("你自己找")。 - AI 反复犯同样的错误,或总是用错版本。 - 用户说"你不是应该记得吗?""上次不是已经验证过了吗?""哪个才是最终版?""别重新做,之前已经跑通了"——这些是记忆与治理边界场景,仅靠记忆不可靠。 - 开始一个新的 AI 辅助项目,希望从第一天起就让 Agent 遵循稳定协议。 - 把一个 AI Agent 接入一个没有规则、错误日志、参数注册表的既有项目。 - 项目已经变乱:文件散落、参数改了没记录、过去的错误反复出现。 - 想强制落地一些持久规则,例如"先查索引再找文件""先计划后执行""文件存在 ≠ 文件有效""参数选择以注册表为准"。 ### 何时不用 - 一次性问答或小改动,不需要项目级约定。 - 项目已有成熟的治理体系,只需要微调某条规则——直接改现有治理文件即可。 ## 记忆与治理的边界 平台记忆(如 Trae 的用户画像 / 项目记忆)是**上下文来源,不是权威事实库**。治理文件才是**项目执行规范**。两者互补: | 层 | 回答什么 | |---|---| | 平台记忆 | "以前发生过什么 / 这个用户通常怎么工作" | | 治理文件 | "这个项目现在必须怎么工作、文件在哪里、哪个版本是权威" | 冲突时的权威优先级: 1. 当前项目文件 / 冻结版本 2. 项目治理文件(`AGENTS.md`、`index.md`、`VERSIONS.md`、注册表) 3. 项目记忆 4. 用户长期记忆 5. AI 推测 当记忆与治理文件冲突时,**治理文件优先**。从记忆中学到的持久约定,必须经人工确认后沉淀进治理文件——记忆本身永远不会成为项目的权威。 ## 使用步骤 ### 第 1 步 —— 初始化 ```bash python scripts/governance.py init --project-dir /path/to/project --project-name "My Project" ``` 从 `templates/` 生成 11 个治理文件(除非传 `--force`,否则不会覆盖已有文件)。 ### 第 2 步 —— 定制 编辑生成的 `AGENTS.md`:目录权限分区、自主权等级、产物存放规则,以及"项目定制"下的项目专属规则。通用核心治理规则保持原样。 ### 第 3 步 —— 日常维护(每个会话) 1. **会话开始**:读 `index.md` → `session_handoff.md` → `LESSONS.md`;生成参数前,读 `blacklist.json` / `whitelist.json`。 2. **工作中**:通过索引找文件(禁止盲目搜索);优先继承 `score > 0.85` 的 whitelist 条目;绝不使用 `permanent_ban: true` 的参数;新错误记入 `LESSONS.md`。 3. **会话结束**:更新 `session_handoff.md`、`index.md`(文件变动)、`CHANGELOG.md`(决策)。 ### 第 4 步 —— 校验、索引、检查 ```bash python scripts/governance.py validate --project-dir /path/to/project # 注册表是否符合 schema python scripts/governance.py index --project-dir /path/to/project # 重建 index.md 目录地图(链接 + 备注) python scripts/governance.py check --project-dir /path/to/project # 健康门禁:文件 + 注册表 + 索引新鲜度 ``` `index` 只重写 `## Root layout` 的机器区块,并更新头部 `Record time`(UTC)。 如果 `## Root layout` 与 `## Change log` 之间检测到人工内容(如自定义 section), `index` 会默认停止并提示;确认要覆盖请加 `--force`(仅解除人工内容保护, 不改变其他生成逻辑)。`check` 在索引过期时会输出 Expected/Actual 差异清单。 ## 输入 / 输出 - **输入**:一个项目目录(有无治理文件皆可)、一个项目名称、用户的治理痛点(文件乱、版本错、反复犯错、"你自己找")。 - **输出**:一个治理工作区(`AGENTS.md`、`index.md`、`VERSIONS.md`、`LESSONS.md`、`session_handoff.md`、`CHANGELOG.md`、`whitelist.json` / `blacklist.json`),以及一个通过校验、索引最新的目录地图。 ## 失败处理 - `init` 失败(路径无效、权限不足):报告失败的具体命令与原因;不要部分初始化或猜测。 - `validate` 报 schema 错误:修复注册表条目;绝不绕过校验。 - `check` 失败(缺文件 / 索引过期):先运行 `index`,再重跑 `check`;仍失败则报告给人类。 - 绝不伪造"通过"结果——如实报告哪些已验证、哪些没有。 ## 产出物 | 文件 | 用途 | |---|---| | `AGENTS.md` | 项目协议:核心治理规则、权威等级、首次运行协议、权限分区、信任边界 |
From trae-community/trae-skills, MIT-0.
More from trae-community/trae-skills
- 0.610score
cycle-delivery
trae-community
Governs evidence-gated software delivery through the local Cycle control plane.
SkillBFreshness: B★ 21Claude Code +9 - 0.607score
entropy-box-zh
trae-community
以箱熵(Entropy Box)的方案咨询 Consult 为核心,把边界明确的具身智能技术需求转化为候选实现方法和有依据的工程工作流。先由当前大模型澄清并拆分宽泛需求,分别咨询具体子问题,再用 Search 深入了解已选技术、用 Lookup 锚定实体、用 Evidence…
SkillBFreshness: B★ 21Claude Code +9 - 0.605score
docx-diff-comment
trae-community
Compare two Word documents to find new features in V2, add comments to the V2 document for each new feature, and generate a requirements table with effort…
SkillBFreshness: B★ 21Claude Code +9 - 0.605score
gbr-pair
trae-community
Pair a phone running Build Remote Agent to this Trae desktop session via gbr/1.
SkillBFreshness: B★ 21Claude Code +9 - 0.605score
cloudbase
trae-community
Use when building, deploying, or debugging Tencent CloudBase (腾讯云开发 / TCB) apps in Trae — Web, WeChat Mini Program, cloud functions, CloudRun, auth…
SkillBFreshness: B★ 21Claude Code +9 - 0.489score
cn-punctuation-checker
trae-community
Checks Chinese text for incorrect English punctuation marks.
SkillDFreshness: D★ 21Claude Code +9 - 0.489score
wechat-mini-program-development
trae-community
WeChat mini-program development skill with standard project structure, request wrapper, and API management.
SkillDFreshness: D★ 21Claude Code +9 - 0.489score
video-to-keyframes
trae-community
Extracts video frames, detects cuts/segments, selects candidate keyframes, and generates review HTML galleries.
SkillDFreshness: D★ 21Claude Code +9 - 0.486score
web-design-teroop
trae-community
A design-first architect that generates, persists, and maintains a comprehensive design document (Style, Color, Typography, Icons, Logo) to ensure a premium…
SkillDFreshness: D★ 21Claude Code +9 - 0.486score
trae-claw-install
trae-community
Run a repository-driven OpenClaw deployment workflow with platform routing, setup/start/check, acceptance checks, and unified troubleshooting.
SkillDFreshness: D★ 21Claude Code +9 - 0.486score
daily-hot-news
trae-community
获取今日全网新闻热榜数据,聚合多个主流平台(微博热搜、百度热搜、知乎热榜、头条热榜、 哔哩哔哩热门、抖音热搜等)的实时热点话题。当用户提到"今日热搜"、"新闻热榜"、 "今天有什么热点"、"全网热搜"、"热门新闻"、"今日新闻"、"热榜"等关键词时触发此技能。 不适用于查询历史新闻或特定领域深度分析。
SkillDFreshness: D★ 21Claude Code +9 - 0.486score
git-commit-generator
trae-community
Generate clear, standardized git commit messages based on code changes (diffs), following Conventional Commits specification.
SkillDFreshness: D★ 21Claude Code +9
Signals
- Stars
- 21
- Forks
- 23
- Open issues
- 1
- Last commit (this folder)
- Sep 10, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT-0
- Repository
- github.com/trae-community/trae-skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from trae-community
- 0.610score
cycle-delivery
trae-community
Governs evidence-gated software delivery through the local Cycle control plane.
SkillBFreshness: B★ 21Claude Code +9 - 0.607score
entropy-box-zh
trae-community
以箱熵(Entropy Box)的方案咨询 Consult 为核心,把边界明确的具身智能技术需求转化为候选实现方法和有依据的工程工作流。先由当前大模型澄清并拆分宽泛需求,分别咨询具体子问题,再用 Search 深入了解已选技术、用 Lookup 锚定实体、用 Evidence…
SkillBFreshness: B★ 21Claude Code +9 - 0.605score
cloudbase
trae-community
Use when building, deploying, or debugging Tencent CloudBase (腾讯云开发 / TCB) apps in Trae — Web, WeChat Mini Program, cloud functions, CloudRun, auth…
SkillBFreshness: B★ 21Claude Code +9 - 0.605score
docx-diff-comment
trae-community
Compare two Word documents to find new features in V2, add comments to the V2 document for each new feature, and generate a requirements table with effort…
SkillBFreshness: B★ 21Claude Code +9 - 0.605score
gbr-pair
trae-community
Pair a phone running Build Remote Agent to this Trae desktop session via gbr/1.
SkillBFreshness: B★ 21Claude Code +9 - 0.489score
video-to-keyframes
trae-community
Extracts video frames, detects cuts/segments, selects candidate keyframes, and generates review HTML galleries.
SkillDFreshness: D★ 21Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get project-governance running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
project-governance skill. Agent Mart. https://agentmart.dev/skills/trae-community-project-governance (last updated 2026-10-24)