project-health-check
Skill1 starsChanged 4 months agoApache-2.0View on GitHub
跨项目只读健康巡检。用于用户明确要求检查多个项目、仓库、分支或 worktree 的状态与遗留工作;先复用项目维护的健康技能,再对未覆盖的项目做有界检查。不自动扫描主目录,不执行清理,也不把普通修复强制升级为治理流程。
6 other projects use a version of this.
Adopted by
Most recently taken up by
- claude-world/director-mode-litetheir own version, first seen Nov 1, 2026
- juandelossantos/another-agent-skillstheir own version, first seen Oct 17, 2026
- dss-time/repo-doctor-skillstheir own version, first seen Oct 15, 2026
- ilamanov/skillstheir own version, first seen Oct 13, 2026
- pnp/sharepoint-skillstheir own version, first seen Aug 24, 2026
- xmqywx/claude-code-skillstheir own version, first seen Apr 2, 2026
Other versions6 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 6
- pnp/sharepoint-skills★ 142same name
- claude-world/director-mode-lite★ 80same name
- xmqywx/claude-code-skills★ 2same name
- dss-time/repo-doctor-skills★ 1same name
- ilamanov/skills★ 1same name
- juandelossantos/another-agent-skills★ 1same 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 repository11 other files it gives its agents.
- 365-chatgpt-pro-deliverySkill
- 365-five-step-dev-claudeSkill
- 365-five-step-dev-codexSkill
- 365-five-step-retro-claudeSkill
- 365-session-rotation-maintainerSkill
- discover-project-skillsSkill
- gws-weekly-digestSkill
- gws-workspaceSkill
- parallel-sessions-protocolSkill
- shopify-store-launch-baselineSkill
- shopify-theme-deliverySkill
The file
Files
项目健康巡检(只读)
检查用户所问的范围,区分事实、推断和未检查项;发现问题不等于获得修复授权。 单个项目存在维护中的健康技能时,先走该项目入口,不叠加本技能的第二套流程。 没有维护中的项目入口时,使用下面的有界后备流程,不猜测不存在的脚本或引用。
确定范围
- 单项目从当前目录或用户指定的项目开始;跨项目从明确给出的路径或应用已保存的
项目清单选择。没有范围且无法可靠确定时,问一个范围问题;不默认递归扫描
$HOME、Library、下载目录或其他无关目录。 - 用户确实请求目录发现时,只在指定根目录内有界发现,兼容
.git文件与目录, 不跟随任意符号链接。先列候选,不自动把发现的每个仓库都纳入深度检查。 - 用 Git common directory / worktree 元数据识别关联;同 remote 的独立克隆仍分别 保留身份。新提交、目录名或最新副本都不能确立 canonical 权威。
- 提交日期不等于任务活动时间。不能仅凭几天没提交判定无主、烂尾或可删除。 报告本次实际检查与不可达/未检查的范围,不把遗漏算作健康。
- 领地边界:worktree 挂靠在其它代理/工具的工作目录(如
~/.codex/)或系统 临时目录下时,视为他方会话的领地——只登记路径、分支与属主线索,不深挖其 index/暂存/改动明细,不列为清理候选;处置建议一律指向属主会话或用户。 巡检范围内的仓库把 worktree 放在哪都够得到,但「够得到」不等于「有权判读」。
按问题选择检查
| 问题 | 最小事实 | 判读限制 |
| 本地工作状态 | root、branch、status、实际 upstream 与 ahead/behind | 缓存 ref 不是远端实时证明;无 upstream 不是零待推 |
| 关联 worktree | common dir、列表、锁定/缺失元数据、必要的 scoped status | dirty/untracked 或旧分支可能是正常工作,不自动清理 |
| PR / CI | 对应仓库、PR head、同 workflow/branch 的运行结果与时间 | 不混算不同流水线;红灯后的新提交不等于修复验证 |
| 待续工作 | 当前模块 checkpoint 或明确相关的任务/回执 | 没有负责人信息就标未知,不读完整历史找罪证 |
| 自带健康工具 | 先检查工具的输入、读写及网络边界,再选必要命令 | 名称带 audit/health 不保证只读,不盲跑 package script |
gh 不可用、未认证或远端受限时,完成可用的本地检查,标明远端未验证;
不触发登录/OAuth、读取凭据或代替用户重跑 CI。需要刷新远端时先明确其必要性,
说明 fetch 会更新本地 refs;不能把缓存结果包装成已获取的最新远端。
输出与后续
按项目简述状态、决定性证据、负责人(或未知)、下一步和覆盖限制。
采用 HEALTHY / ATTENTION / ACTION_REQUIRED / BLOCKED,仅对已检查范围下结论;
缺失数据标为 UNAVAILABLE / NOT_CHECKED。年龄/未推提交只是背景,是否阻塞取决于
用户所问的下一步。小问题无需固定一页报告、A/B/C 分级或附完整命令输出。
- 不输出原始 remote URL、私有地址、凭据、环境文件、原始个人信息或未净化的日志。
- 不自动改文件、reset、stash、clean、删分支/worktree、创建/归档任务、ACK、改绑定、 push、合并、重跑 CI、发布或发消息;建议不授予这些操作。
- 用户已明确要求某项修复时,转目标项目的具体实现技能,在该授权范围内继续; 普通修复不强制五步法、另建任务或重复确认。实际受保护动作仍遵守其操作器与授权。
- 不因一次巡检自行设置定时器;已有监控按产品机制执行,状态无变化时保持安静。
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.

