feel
Pass
Audited by Gen Agent Trust Hub on Apr 11, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: Employs shell utilities like
git,gh, andjqto compute activity metrics from the local environment. - [DATA_EXFILTRATION]: Performs a read-only count of session files in the application's local project directory to determine usage frequency. No data is transmitted externally.
- [PROMPT_INJECTION]: Contains a potential surface for indirect prompt injection via the processing of Git logs and GitHub issue data.
- Ingestion points: Git commit messages and GitHub issue metadata.
- Boundary markers: None.
- Capability inventory: Read-only access to repository data and write access to a local log file.
- Sanitization: None, but the skill is safe as it focuses on metadata extraction (line counts) rather than content execution.
Audit Metadata