product-capability
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains only markdown instructions and no executable scripts or system commands. It follows standard prompt engineering practices for structured output generation.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from product documents and discussions. 1. Ingestion points: User-provided product intent and repository documentation (SKILL.md). 2. Boundary markers: Not present. 3. Capability inventory: Structured text generation only; no sensitive file access or network operations across any files. 4. Sanitization: Not present. Although this represents an attack surface for indirect prompt injection, the risk is negligible as the skill only produces text-based artifacts and does not invoke sensitive tools.
Audit Metadata