skill-cleaner
Pass
Audited by Gen Agent Trust Hub on Jun 12, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill is designed for local text manipulation and structural improvements of skill definitions within the repository.
- [COMMAND_EXECUTION]: The skill instructs the agent to use
git log -pto understand the history of a target skill before editing, which is a standard local development practice used here to ensure context from previous fixes is preserved. - [DATA_EXPOSURE]: The skill accesses local file paths within the repository's
skills/directory for the purpose of editing. It includes a safety check that explicitly prohibits editing external configuration directories like.agents/skills/or.claude/skills/to prevent accidental overwriting of managed meta-skills.
Audit Metadata