Improve/skill review optimization

[yogesh-tessl]

Hey @armelhbobdad 👋

"Every instruction links back to a specific file and line in the source it was compiled from". That's the right answer to the AI hallucination problem. The AST-traced citation chain from source to skill is the kind of rigor this space needs. Impressive that you've built this as a full BMAD module with tiered verification profiles across five languages. Wanted to suggest a few improvements to the SKILL.md.

I ran your skills through tessl skill review at work and found some targeted improvements. Here's the full before/after:

Skill	Before	After	Change
skf-forger	48%	90%	+42%
skf-test-skill	64%	88%	+24%
skf-create-skill	78%	88%	+10%
skf-analyze-source	78%	88%	+10%
skf-audit-skill	84%	94%	+10%

What changed

All 5 skills - description improvements:

• Replaced vague/metaphorical language with concrete action descriptions listing specific capabilities
• Added broader natural trigger terms to Use when clauses (e.g. "build a skill", "check skill accuracy", "find skills in codebase") so agents match on more natural user phrasings
• Ensured descriptions use quoted string format in frontmatter

skf-forger (biggest improvement, +42%):

• Rewrote description from vague "forge master" metaphor to concrete lifecycle actions (analyzes repos, compiles briefs, audits drift, tests completeness, exports packages)
• Condensed Identity & Principles and Communication Style sections into a single concise Principles block
• Trimmed the Overview to focus on what Ferris does rather than how it's described

skf-test-skill (+24%):

• Replaced abstract jargon ("cognitive completeness verification") with concrete actions (checks API coverage, validates coherence, produces scored PASS/FAIL report)
• Added trigger terms: "check skill quality", "validate skill", "is my skill ready"
• Removed redundant Role section

skf-create-skill, skf-analyze-source, skf-audit-skill (+10% each):

• Expanded descriptions with specific actions and additional trigger terms
• Trimmed verbose Overview and Role sections

I kept this PR focused on the 5 skills with the biggest improvements to keep the diff reviewable. Happy to follow up with the rest in a separate PR if you'd like.

Honest disclosure. I work at https://github.com/tesslio where we build tooling around skills like these. Not a pitch - just saw room for improvement and wanted to contribute.

I also added a lightweight GitHub Action that auto-reviews any skill.md changed in a PR (includes min permissions, uses a pinned action version, only posts a review comment).

This means that it gives you and your contributors an instant quality signal before you have to review yourself (no signup, no tokens needed).

Want to self-improve your skills? Just point your agent (Claude Code, Codex, etc.) at this Tessl guide (https://docs.tessl.io/evaluate/optimize-a-skill-using-best-practices) and ask it to optimize your skill. Ping me - @yogesh-tessl (https://github.com/yogesh-tessl) - if you hit any snags.

Thanks in advance 🙏