audit-skill
audit-skill is a Skill evaluation & testing skill for AI agents, published by jellyrock in the jellyrock repository on GitHub. Skillspace rates audit-skill D (59/100) and it passed the safety scan; the rating is based on static analysis only.
Audit a specific skill's recent execution against the current Claude Code session transcript. Runs the mechanical extractor at .claude/skills/audit-skill/extract-friction.cjs which produces JSON across four dimensions — friction findings (repeated-command, failed-recovery, confusion-marker, lint-spam, changelog-edit-attempt, tasks-leakage, test-claim-without-evidence, hardware-claim-mismatch, permission-gap), performance (clock time, token usage per model, cache-hit ratio, cost estimate), model-fit profile, and the auditor's judgment of output accuracy (the only dimension the extractor cannot mechanize). The skill prose classifies findings, weighs perf anomalies, proposes concrete edits to the audited skill's SKILL.md or supporting infra, and assesses whether the skill's `model:` setting still fits. Use after any non-trivial skill invocation — friction-clean runs can still be slow, token-bloated, or trigger permission prompts that should be allowlisted.
- Grade
- D
- Quality
- 59 / 100
- Safety
- Passed
- GitHub stars
- 35
Unproven: rated by static analysis only, not yet compared head-to-head.
This rating describes the version of the repository we crawled; it may have changed since, so go by its current files.
Our review
- Architecture
- 80 / 100
- Maturity
- Structured: ships references or scripts
- Security scan
- Low-risk patterns
- License
- GPL-2.0
- Adoption
- Adopted by 0 other repositories, in 0 collections
What the scanner flagged
- mentions secrets in text
Not checked yet
- Category is a classifier prediction; no model has read this skill yet
- Not judged on the rubric yet (not a contender in its category)
- No blind task eval yet
- Scope (portable / vendor / repo-internal) not assessed; only the top of each category is read for scope
- Sibling files (scripts, references) not fetched; security scan covers SKILL.md only
Skillspace indexes and rates this skill; it does not host or endorse it. Run third-party skills at your own risk.