thedaviddias

thedaviddias / skill-check

Public

Linter for agent skill files

ai-agents checker linter skill skillmd
58
2
100% credibility
Found Feb 23, 2026 at 24 stars 2x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

skill-check validates AI agent skill files by checking structure, quality, length limits, duplicates, links, and security risks with scores, auto-fixes, and various report formats.

How It Works

1
🕵️ Discover skill-check

While building AI helpers, you learn about a friendly checker that ensures your skill instructions are clear and safe.

2
⚙️ Try it on your folder

Point it at your skills folder and run a simple check to see how your instructions measure up.

3
📊 See scores and tips

Beautiful reports pop up with quality scores, issue highlights, and smart suggestions to improve.

4
🔧 Fix issues easily

Use one-click fixes for simple problems or follow tips to polish your skills manually.

5
🛡️ Scan for safety

Quickly check for any security worries in your skills to keep everything protected.

6
📱 Share or integrate

Generate shareable cards, reports, or add it to your workflow for ongoing checks.

✅ Skills ready to shine

Your AI skills now have top scores, no issues, and are secure—ready for agents to use confidently!

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 24 to 58 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-check?

skill-check is a TypeScript CLI linter for agent skill files like SKILL.md, validating frontmatter, descriptions, body limits, links, and duplicates while computing 0-100 quality scores. It catches issues in AI agent skill directories, auto-fixes safe problems like missing metadata or formatting, and bundles security scanning via agent-scan. Run `npx skill-check .` for instant diagnostics, or pipe into GitHub Actions for PR annotations.

Why is it gaining traction?

Unlike generic markdown linters, it understands agent skill structure—scoring descriptions for "Use when" phrasing, splitting oversized bodies into references, and detecting duplicates so agents pick the right tool. GitHub Action support with SARIF upload and remote repo scanning (no clone needed) hooks CI/CD users, while `--fix --interactive` and watch mode speed up iteration. Quality scores and shareable cards make tracking improvements addictive.

Who should use this?

AI agent builders maintaining skills folders for Claude or Cursor, especially teams enforcing consistency across dozens of SKILL.md files. Devs integrating linter GitHub Actions on pull requests, or solo prototypers using `skill-check watch` during skill development. Perfect for open-source repos sharing agent skills, dodging vague descriptions or bloated bodies.

Verdict

Grab it for agent skill linting—solid docs, CLI polish, and CI integration punch above 18 stars, though 1.0% credibility signals early maturity; test coverage hits 80% thresholds. Use in CI now, contribute rules as your workflow evolves.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.