repo-health
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates as a read-only audit tool by default, retrieving repository metadata through standard git and GitHub CLI commands. It includes specific safeguards against automated modification of files and prohibits the execution of installation commands.
- [DATA_EXFILTRATION]: The tool includes a security check to detect if sensitive files (e.g., .env, credentials) are tracked in the repository's history, which helps prevent accidental credential exposure. The check is performed locally and does not involve exfiltration of data.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the repository being audited, including file names, README.md contents, and GitHub API responses. Ingestion points: File presence checks (Step 3), GitHub metadata (Step 4), and README scanning (Step 5). Boundary markers: Absent, however, the skill's logic is constrained to specific keyword matching. Capability inventory: The skill can execute local audit commands and write new template files if confirmed by the user. Sanitization: Step 8 explicitly requires an 'infra/PII scrub' on all generated content before it is written to the file system.
Audit Metadata