writing-agents-md
Pass
Audited by Gen Agent Trust Hub on Jul 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill analyzes untrusted repository content (e.g., README.md, package.json) to generate or review documentation, creating a surface for indirect prompt injection.\n
- Ingestion points: The skill (Grounding workflow section) identifies repository files and manifests as data sources for agent inspection.\n
- Boundary markers: The skill does not specify the use of delimiters or instructions to ignore embedded commands when reading files.\n
- Capability inventory: The skill is restricted to text generation and review, with no direct execution or network capabilities.\n
- Sanitization: The instructions explicitly forbid including secrets or security-bypassing commands in the output.
Audit Metadata