Thanks for your interest in improving this skill.
If the skill produces unexpected output, misscores content, or breaks during the flow:
- Open an issue
- Include:
- Your Claude Code version (
claude --version) - What step of the audit broke or behaved unexpectedly
- What you expected vs. what happened
- The content you were auditing (or a sanitized version)
- Your Claude Code version (
Ideas for new scoring layers, better rubric criteria, or flow improvements:
- Open an issue describing the change
- Explain why it improves the audit for the end user
- If proposing a scoring change, include a before/after example
- Fork the repo
- Create a branch (
git checkout -b improve-scoring-rubric) - Make your changes
- Test by running
/whystrohm-auditin a fresh Claude Code session - Submit a PR with a clear description of what changed and why
See CODE_OF_CONDUCT.md.
This skill is maintained by WhyStrohm. For questions about the methodology or the full content infrastructure service, visit https://whystrohm.com/audit.