api-design
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to study external content (live URLs, Figma, source code) to guide API design, creating an ingestion surface for potential indirect prompt injection.
- Ingestion points: Live URLs, Figma pages, and frontend source code mentioned in the Consumer-Driven Design section.
- Boundary markers: No explicit markers or instructions to disregard embedded commands in the external data are present.
- Capability inventory: The agent context usually includes tools for web browsing and file system access.
- Sanitization: No validation or sanitization steps are provided for the ingested data.
Audit Metadata