Laohan Skillsdoctor — Development skill for Claude Code
Claude Code / OpenClaw skill 体检仪表盘 — 自动来源分类/污染副本/漂移/孤儿检测.
How to install Laohan Skillsdoctor
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open hanzhcn/laohan-skillsdoctor and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Laohan Skillsdoctor does
Claude Code / OpenClaw skill 体检仪表盘 — 自动来源分类/污染副本/漂移/孤儿检测.
Alternatives in Development
- Openclaw China — OpenClaw中国插件:支持飞书,钉钉,QQ,企业微信,微信 4k ★
- Antigravity Claude Proxy — Proxy that exposes Antigravity provided claude / gemini models, so we can use them in Claude Code and OpenClaw 3.5k ★
- Agentic Stack — One brain, many harnesses 2.2k ★
README
laohan-skillsdoctor
原名 skill-doctor,2026-10 起更名为 laohan-skillsdoctor。
Claude Code / OpenClaw **skill 体检仪表盘** —— 一目了然看清你的 skills 来自哪里、有没有污染/漂移/孤儿。
[English](./README.en.md) · 简体中文
为什么做
用 Claude Code / OpenClaw 一段时间后,`~/.agents/skills` 和 `~/.claude/skills` 会堆出几百个 skill,来自四面八方:自己写的、`npx skills add` 装的、ECC 等插件自带的、OpenClaw 自带的……还会被某次操作批量塞进一堆**插件 skill 副本**(污染),或者自研 skill 的本地副本和 git 仓库**双向漂移**(你改了本地没推、仓库改了没拉,CC 一直读旧版)。
现成的 skill 管理器(jiweiyeah/Skills-Manager 等)都是"入库分发"或"手动打标签",**没有一个自动按来源分类、检测污染/漂移/孤儿**。laohan-skillsdoctor 把这套审计逻辑产品化。
功能
- 自动来源分类:自研(laohan/cheat)/ npx / 插件副本 / OpenClaw / 孤儿,一眼看清
- 健康检查:
- 🔴 失效链接(dangling symlink)
- 🔴 CC 不可见(agents 有但 claude 没)
- 🟡 插件副本(和插件 skill 同名,可能是污染)
- 🟡 双份实体(agents + claude 都是实体,漂移温床)
- 🟡 漂移(自研 skill 的本地 vs git 仓库 diff)
- 详情视图:路径状态、npx 锁文件元数据、插件原件路径、漂移 diff、SKILL.md 预览
安装
git clone https://github.com/hanzhcn/laohan-skillsdoctor.git
cd laohan-skillsdoctor
npm install
npm run dev
# 打开 http://localhost:3000
纯本地运行,只读扫描你的 skills 目录,不修改任何文件。
数据源
| 来源 | 路径 |
|---|---|
| agents 真身 | ~/.agents/skills/ |
| CC 桥接层 | ~/.claude/skills/ |
| npx 锁文件 | ~/.agents/.skill-lock.json |
| 插件清单 | ~/.claude/plugins/installed_plugins.json |
| 自研 git | ~/Documents/laohan-skills/ |
| cheat git | ~/Documents/cheat-on-content/skills/ |
| OpenClaw | ~/.openclaw/skills/ |
分类规则
- symlink 指向
Documents/laohan-skills→ 自研 laohan - symlink 指向
cheat-on-content→ 自研 cheat - symlink 指向
.openclaw/skills→ OpenClaw .skill-lock.json有记录 → npx- 与某插件 skills/ 同名 → 插件副本
- 以上都不是 → 孤儿 ⚠️
技术栈
Next.js 16 (App Router) · React 19 · TypeScript · Tailwind 4 · shadcn/ui
路线图
- 只读体检(扫描/分类/健康检查/详情)
- 写操作(删除副本/转 symlink/同步仓库,带二次确认)
- 更新检测(git fetch / skillFolderHash 比对)
- Tauri 桌面打包
License
MIT
Related Skills
Nexting
Remote control for Claude Code, Codex, Grok, and Cursor on Mac or PC. View sessions, send tasks, and drive the
Opik Openclaw
🦞 Official plugin for OpenClaw that exports agent traces to Opik. See and monitor agent behaviour, cost, toke
AgentHandover
What if OpenClaw, Claude Code, Codex etc. knew how to do your work without you saying it? AgentHandover observ
Phistory
Phistory automatically archives versioned system prompt snapshots from agent CLIs like Claude Code, Codex, Ope
Openclaw Billing Proxy
Route OpenClaw API requests through Claude Code subscription billing instead of Extra Usage
Agnes Help Skill
这是一个专门接入Agnes免费模型API的skill,帮助新手排查和修复问题,也能直接在Agent工具中使用生图生视频,支持Claude code、Openclaw、Codex、Hermes等工具部署接入,所有信息基于最
Related Agents
ClawPort -- Setup Guide
This guide walks you through getting ClawPort running against your own OpenClaw instance. If you just want the
Part 16: autoDream — Automatic Memory Consolidation
Note: OpenClaw 2026.4+ has a built-in dreaming system (Part 22) inside memory-core. If you're on 2026.4+, use
CyberClaw
👾 下一代透明智能体架构 Next-Gen Transparent Agent Architecture 🔍 全行为审计 🛡️ 两段式安全调用 🧠 双水位记忆 ⏰ 心跳任务 📊 P0 级事故率降低 80% 兼容