critique-method
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external data from DESIGN.md, PRODUCT.md, and various UI source files. This creates an entry point for indirect prompt injection if those files contain malicious instructions intended to influence the agent's critique output.
- Ingestion points: SKILL.md (via instructions to read DESIGN.md, PRODUCT.md, and scan UI files).
- Boundary markers: None explicitly defined to separate data from instructions.
- Capability inventory: File system read access (implicit in the use of grep and context reading).
- Sanitization: No sanitization or validation of the input file content is mentioned.
Audit Metadata