lets-build
Pass
Audited by Gen Agent Trust Hub on Nov 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes external ticket or task content, which could include hidden instructions.
- Ingestion points: Written tickets or tasks identified by ID, URL, file path, or text (SKILL.md).
- Boundary markers: There are no explicit markers or 'ignore' instructions for the processed content.
- Capability inventory: The skill has access to building capabilities (via /build-ticket) and ticket state modification (via /to-tickets).
- Sanitization: No specific sanitization or filtering of input data is defined, although the user-in-the-loop design phase serves as a manual checkpoint.
Audit Metadata