claude-skills-benchmark
Installation
SKILL.md
Skill Benchmarking
Evaluate Agent Skills through static analysis and evaluation-driven methodology. Source: Anthropic's skill evaluation guidance.
When to Use
Activate when:
- Assessing skill quality across a plugin or marketplace
- Measuring skill activation accuracy (false positives/negatives)
- Comparing skill versions or skill-vs-no-skill performance
- Running the
/benchmark-skillscommand - Reviewing skill descriptions for optimization
Static Analysis Checks
Run these checks against every skill to produce a quality scorecard: