Plugin self eval loop
Skill patrykkopycinski/patryks-treadmill-claude-plugins/skills/plugin-self-eval-loop
Self-evaluation loop for the treadmill Claude plugins pack. Runs cursor-plugin-evals against bundled skills, tracks quality over time, and escalates recurring failures via PAMS. Use periodically or before publishing plugin updates.From its SKILL.md
npx -y skills add patrykkopycinski/patryks-treadmill-claude-plugins --skill plugin-self-eval-loopAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 3 stars3 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
1.2 KB, as published. Nobody here has run it
Plugin Self-Eval Loop
When to Use
- Before publishing any plugin update
- Periodic quality check (monthly recommended)
- After adding or modifying skills in the pack
- When PAMS detects recurring correction patterns in skills from this pack
Process
- Run evals — execute
cursor-plugin-evals skill-evalagainst each skill in the pack - Parse results — identify failing skills and evaluators
- Fix — for each failing skill, apply targeted fix (one category per iteration)
- Re-run — re-evaluate only the failing skills
- Iterate — max 3 iterations per skill
- PAMS escalation — if a skill fails repeatedly (3+ sessions), escalate:
- 2x → create a rule documenting the anti-pattern
- 3x → create a new skill + eval to address the gap
Convergence
- All skills pass at >= 85% (OSS-equivalent threshold for plugin skills)
- No regressions in previously passing skills
- PAMS escalation triggers acted on