xigua-wang

xigua-wang / skill-doctor

Public

Local-first inspector for coding-agent skills, conflicts, precedence, and risk analysis.

17
0
100% credibility
Found Apr 20, 2026 at 17 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Skill Doctor is a local inspector that scans AI coding agent skill directories for conflicts, risks, and overrides, displaying results in a bilingual web dashboard with optional AI-enhanced analysis.

How It Works

1
🕵️‍♀️ Discover Skill Doctor

You hear about Skill Doctor from a coding friend, a simple tool to peek inside your AI helpers' skill folders.

2
🚀 Start the inspector

Open it right from your project folder, and a friendly web page pops up in your browser.

3
🔍 Scan your workspace

Pick your project folder and hit scan to uncover all the skills your AI helpers have installed.

4
📊 See the overview

The dashboard shows totals for skills, conflicts, high risks, and folders checked across your setup.

5
👁️ Dive into details

Click a skill to reveal its triggers, files, risks, and which one wins in overlaps.

6
🧠 Unlock smart advice

Link an AI thinker for quick summaries, key findings, and helpful suggestions on fixes.

✅ Skills under control

You now clearly see what your AI helpers do, spot dangers early, and keep your workspace safe and predictable.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 17 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is skill-doctor?

Skill Doctor is a local-first inspector for coding-agent skills across project, global, and system scopes. It scans directories like OpenClaw's skills folders or Cursor's .cursor/skills, flags conflicts, precedence chains, trigger overlaps, and risks like shell exec or network calls, then layers on AI analysis via OpenAI-compatible APIs. Built in TypeScript with a React UI and CLI (npx skill-doctor --scan), it stores immutable snapshots in ~/.skill-doctor/ for history and comparison.

Why is it gaining traction?

Unlike cloud-based agent dashboards, it's fully local-first on GitHub workflows—no accounts, instant scans via skill-doctor-scan --markdown report.md. Devs dig the bilingual UI showing winner skills in precedence chains, high-risk patterns, and model-powered spotlights, plus OpenClaw extras from openclaw.json. It turns hidden prompt behaviors into inspectable infrastructure, surfacing surprises before they hit production.

Who should use this?

AI agent users building coding-agent setups with OpenClaw, Cursor, Copilot, or Claude, especially those stacking project skills over globals and chasing skill check doctor builds for conflicts or risk analysis. Ideal for teams auditing precedence in multi-root workspaces or solo devs wanting quick CLI reports on destructive patterns before deploying.

Verdict

Try it for local-first skill doctor analysis if you're deep in agent skills—CLI and UI deliver real value out of the box. At 17 stars and 1.0% credibility, it's early-stage with solid bilingual docs but no license yet; mature it with per-agent rules for broader adoption.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.