api-design
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates on untrusted external inputs such as feature requirements and existing route handlers, creating a potential vector for indirect instructions.
- Ingestion points: Existing route handlers, schema design output, and feature requirements (SKILL.md, Procedure section).
- Boundary markers: None. The skill does not define specific delimiters (e.g., XML tags or triple quotes) to separate ingested data from agent instructions.
- Capability inventory: The skill reads local project files and generates documentation. It does not invoke shell commands or perform network requests.
- Sanitization: No sanitization or filtering logic is present in the instructions to handle malicious content within the analyzed files.
Audit Metadata