design
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from design documentation and application source code to perform its core functions, creating an attack surface for indirect prompt injection.\n
- Ingestion points: The skill reads and treats user-provided bullets in
docs/design.mdas the source of truth and inventories route behavior from existing application source code (SKILL.md, step 2 and Initialization).\n - Boundary markers: There are no explicit delimiters or boundary markers used to separate the ingested documentation or source code from the agent's internal instructions.\n
- Capability inventory: The skill is authorized to implement and ship user-facing UI screens, components, and locale files (SKILL.md, Implement UI).\n
- Sanitization: The skill lacks specified validation or sanitization mechanisms for the input data it reads from the repository before using it to inform code generation.
Audit Metadata