Skip to content

Latest commit

 

History

History
40 lines (26 loc) · 1.53 KB

File metadata and controls

40 lines (26 loc) · 1.53 KB

Contributing

Issues and PRs welcome.

Compatibility reports

The most valuable contribution right now is testing the skill on platforms other than Claude. If you test on Gemini CLI, Codex, Copilot, Cursor, or any other agentskills.io-compatible platform, open an issue using the Compatibility Report template with:

  • Platform and version
  • Model used
  • What worked
  • What didn't
  • Any workarounds applied

Making changes

  1. Fork the repo and create a branch.
  2. Make your changes. If you modify files inside skill-provenance/, follow the versioning protocol:
    • Update per-file version integers in MANIFEST.yaml for changed files.
    • Run ./skill-provenance/validate.sh --update to recompute hashes.
    • Add a new top entry to skill-provenance/CHANGELOG.md describing what changed.
    • Mirror the same release entry into root CHANGELOG.md, which is the full archive.
    • Bump bundle_version in MANIFEST.yaml (PATCH for fixes, MINOR for features, MAJOR for breaking changes).
  3. Run ./skill-provenance/validate.sh to confirm everything is clean.
  4. Open a PR with a clear description of what changed and why.

What doesn't need the versioning protocol

Changes to repo-level files that aren't part of the skill bundle (root README.md, AGENTS.md, CONTRIBUTING.md, .github/, LICENSE) don't require manifest updates or version bumps.

Code style

  • SKILL.md and README.md: wrap prose at ~80 characters.
  • YAML: 2-space indent.
  • Bash: set -euo pipefail, zero external dependencies.
  • JSON: 2-space indent.