skill-review
Pass
Audited by Gen Agent Trust Hub on Apr 15, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands to inventory skill files and check git history. These operations are restricted to retrieving local file metadata (
stat,date) and viewing the commit log (git log). - [DATA_EXFILTRATION]: No network operations or external data transfers were identified. The skill only interacts with local files and the local git repository.
- [PROMPT_INJECTION]: The instructions focus entirely on maintenance workflows and do not contain any patterns aimed at bypassing safety guidelines or overriding agent constraints.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface by reading other skill files (
skills/*/SKILL.md) and shared lesson files (shared/gotchas.md). However, these are internal configuration files rather than untrusted external data, posing negligible risk in this context.
Audit Metadata