Thank you for your interest in contributing to the Agent Skills repository! This project aims to provide a collection of specialized skills for AI agents.
- Create a Directory: New skills should be added to the
skills/directory at the repository root. - Follow Conventions: Ensure your skill follows the established structure and naming conventions used by existing skills in
.agents/skills/. - Documentation: Include a
SKILL.mdor similar documentation file explaining the skill's purpose and usage.
- Identify the skill in
skills/or.agents/skills/. - Make your improvements, ensuring they are backward compatible or clearly document breaking changes.
- Ensure the skill remains independent and doesn't introduce unnecessary dependencies.
- Code Quality: Follow the guidelines in
AGENTS.mdfor code generation and review. - Simplicity: Prioritize simplicity and clarity.
- Security: Never commit secrets or sensitive information.
- Testing: Verify your changes thoroughly.
- Message Style: Use concise, descriptive commit messages (no longer than a Tweet, 280 characters). Follow conventional commits
- Scope: Focus each commit on a single logical change.
- Secrets: Double-check that no secrets (API keys, credentials) are included in your commits.
- Verification: Ensure that your changes pass all local checks (linting, tests) before committing.
- Create a new branch for your changes.
- Commit your changes with clear, descriptive messages.
- Open a Pull Request (PR) against the
mainbranch. - Participate in the code review process and address any feedback.
We appreciate your contributions!