Back to browse
GitHub Repository

Evaluate agent skill quality. Find the weakest link. Fix it. Prove it worked.

224 starsJavaScript

SkillCompass – open-source quality evaluator for your AI skills

by yo103jg·Apr 13, 2026·2 points·0 comments

AI Analysis

●●SolidNiche GemBig Brain

Six-dimension scoring for Claude skills when nothing else measures quality this way.

Strengths
  • Usage-driven suggestions track what's stale, unused, or risky across your skill library
  • Auto-evaluation gate catches quality regressions when any tool edits a skill file
  • Version tracking proves improvements worked instead of tweak-and-hope workflows
Weaknesses
  • Only works with Claude Code and OpenClaw, excludes other agent frameworks entirely
  • Requires Claude Opus 4.6 for scoring, adding significant API cost per evaluation
Target Audience

Claude Code users maintaining custom agent skills

Similar Projects