Thank you for considering contributing to the OpenCode Agent Loop. This project aims to provide a reliable, safe, and extensible framework for autonomous software development.
This project adheres to a Code of Conduct. By participating, you agree to maintain a respectful and inclusive environment.
- Check existing issues for duplicates
- Use the bug report template
- Include your configuration (sanitized), steps to reproduce, and logs
- Do not include credentials or secrets in bug reports
- Check existing issues and discussions for similar ideas
- Use the feature request template
- Explain the problem and proposed solution
- Be specific about which agent roles would be affected
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Make your changes
- Run tests:
npm test - Run validation:
bash scripts/validate.sh - Commit with conventional commit messages
- Push and open a Pull Request
git clone <repository-url>
cd opencode-agent-loop
npm install# Full test suite
npm test
# Individual test suites
node tests/routing-tests.mjs
node tests/runtime-tests.mjs
node tests/tool-integration-tests.mjs
node tests/bypass-detection.mjsUse conventional commits format:
feat: add new feature
fix: correct bug
docs: update documentation
chore: maintenance tasks
refactor: code restructuring
test: add or update tests
style: formatting changes
feature/<description>— New featuresfix/<description>— Bug fixesdocs/<description>— Documentationchore/<description>— Maintenance
All submissions require review. The review agent will inspect:
- Correctness against acceptance criteria
- Security implications
- Test coverage
- Documentation accuracy
- Potential regressions
If your contribution was created with assistance from AI tools:
- Disclose this in your pull request
- Review all generated code for correctness and security
- Ensure you have the right to submit the code
- The same review standards apply regardless of origin
Open a GitHub Discussion for questions and community support.