OpenClaw Skill Tester

🧪 CLI tool to test and validate OpenClaw skills - automated quality checks, dependency verification, and testing reports.

Why?

With 5,400+ skills in the OpenClaw ecosystem, it's hard to know which ones work reliably. This tool helps:

• Skill developers: Test your skills before publishing
• Skill users: Verify a skill works before installing
• CI/CD: Automate skill quality checks

Features (Phase 1 - CLI)

• ✅ Validate skill structure (SKILL.md, dependencies)
• ✅ Check for required files and metadata
• ✅ Test skill execution in isolated environment
• ✅ Generate quality reports with scores
• ✅ Support for batch testing multiple skills

Installation

npm install -g openclaw-skill-tester

Package: https://www.npmjs.com/package/openclaw-skill-tester

Usage

Test a Single Skill

openclaw-skill-tester test ./path/to/skill

Batch Testing

Test multiple skills in a directory:

openclaw-skill-tester test ./skills --batch

JSON Output

openclaw-skill-tester test ./skill --report json

CI/CD Integration

Exit with code 1 if tests fail (perfect for CI pipelines):

openclaw-skill-tester test ./skill --ci

GitHub Actions Example

name: Test Skill Quality
on: [push, pull_request]

jobs:
  test:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3
      - uses: actions/setup-node@v3
      - run: npm install -g openclaw-skill-tester
      - run: openclaw-skill-tester test . --ci

Web Interface (Phase 3)

Run the web server locally:

git clone https://github.com/alex-vbcoding/openclaw-skill-tester.git
cd openclaw-skill-tester
npm install
npm run web

Then open http://localhost:3000 to see skill rankings and test results!

Roadmap

Phase 1 (Complete ✅): CLI tool

• [x] Project setup
• [x] Skill structure validation
• [x] Quality scoring (0-100)
• [x] Report generation (text + JSON)
• [x] Batch testing
• [x] CI/CD integration
• [x] Security checks (credentials, dangerous patterns)
• [x] Published to npm v0.1.0

Phase 2 (Complete ✅): Advanced testing

• [x] Basic execution tests (npm install + run)
• [x] Dependency conflict detection
• [x] Security vulnerability scanning (npm audit)
• [x] Performance benchmarks (size, load time)
• [x] Dependency analysis
• [x] --no-execution and --no-performance flags

Phase 3 (Complete ✅): Web interface

• [x] Web server (Express + SQLite)
• [x] Database for test results
• [x] Beautiful responsive UI
• [x] Skill rankings page
• [x] Recent tests display
• [x] Directory upload functionality
• [x] Real-time test execution
• [x] Automatic page refresh with results
• [ ] Community ratings (future enhancement)
• [ ] Public deployment (future)

Contributing

Contributions welcome! This project is maintained by Alex Chen ([email protected]).

License

MIT

Status: 🚧 Active development - first commit $(date +%Y-%m-%d)