test-strategy-author
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted project data to build test strategies. 1. Ingestion points: Reads project documentation (e.g., docs/risks.md) and source code files. 2. Boundary markers: Lacks instructions for explicit delimiters or safety warnings when processing external data. 3. Capability inventory: Intended for generating and updating markdown documentation files; no shell execution or network access patterns are identified. 4. Sanitization: No explicit content filtering or escaping is mandated by the instructions.
Audit Metadata