skill-compass — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited skill-compass (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
A wrapper around the SkillCompass Node tool installed at ~/.claude/skill-compass/. Use this skill when the user wants to:
The tool exposes its slash-commands at ~/.claude/skill-compass/commands/:
/eval-audit # full audit of installed skills, scores each
/eval-compare # head-to-head compare two skills on the same input
/eval-evolve # iteratively improve a skill based on its own audit
/eval-improve # one-shot improve a skill
/eval-merge # merge two overlapping skills into oneFrom a Claude Code session:
node ~/.claude/skill-compass/oc/run.js audit ~/.claude/skills/<skill-name>
node ~/.claude/skill-compass/oc/run.js audit-all ~/.claude/skills/Or read the SkillCompass AGENTS.md and commands/eval-audit.md for the full agent-driven flow:
cat ~/.claude/skill-compass/AGENTS.md
cat ~/.claude/skill-compass/commands/eval-audit.mdthe user has 200+ skills installed. Without this tool, identifying dead weight is manual eyeballing. SkillCompass produces ranked output with pass/fail signals per skill so the audit produces a punch list, not a wall of text.
~/.claude/skill-compass/~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.