Contributing to Research-First Subagent Framework

October 1, 2025 · View on GitHub

Thank you for your interest in contributing! This framework helps developers use Claude Code subagents effectively through the research-first pattern.

Ways to Contribute

If you've successfully used this framework in your project, share your experience:

Create case study in docs/examples/your-project-case-study.md
Include: project scale, agents created, results achieved
Submit as pull request

2. Improve Agent Templates

Enhance or create new agent templates:

Ensure agents are research-only (no implementation)
Include clear capabilities and anti-patterns
Test template before submitting
Submit to docs/agent-templates/

3. Enhance Documentation

Fix typos, clarify concepts
Add examples and code samples
Improve troubleshooting guides
Expand implementation guides

4. Report Issues

Found a problem? Open an issue
Include context: what you tried, what happened, what you expected
For documentation issues: include page/section reference

Contribution Guidelines

Agent Template Guidelines

All research agents must:

✅ Be research-only (explicitly state "NEVER implement code")
✅ Define clear research capabilities
✅ Specify output format and location
✅ Include anti-patterns section
✅ Provide example usage

Documentation Guidelines

✅ Clear, concise writing
✅ Code examples are copy-pasteable
✅ Links work and are relevant
✅ Follows existing documentation structure

Code of Conduct

Be respectful and constructive
Focus on improving the framework
Help others learn the pattern
Assume good intentions

Pull Request Process

Fork the repository
Create a feature branch: git checkout -b feature/your-feature
Make your changes
Test if applicable
Commit with clear messages: feat: add security research agent template
Push to your fork
Open pull request with description of changes

Questions?

Review existing documentation in docs/
Check troubleshooting guide
Open an issue for discussion

Thank you for helping make this framework better!