readme-updater
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [Indirect Prompt Injection] (LOW): The file references/discovery_patterns.md defines patterns for ingesting untrusted data from external project files such as package.json and README.md. This establishes an attack surface for indirect prompt injection. 1. Ingestion points: External metadata and documentation files (e.g., package.json, pyproject.toml). 2. Boundary markers: Not defined in the provided files. 3. Capability inventory: No executable scripts or logic provided in this skill. 4. Sanitization: No sanitization or validation logic is present in the templates.
- [No Code] (SAFE): The skill consists only of markdown reference files and does not contain any executable scripts (.py, .js, .sh) or active instructions for the agent to perform actions.
Audit Metadata