tk-domain
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data such as repository source code, product materials, and user conversations, which could contain adversarial instructions intended to influence the agent's output.\n
- Ingestion points: Repository source, product material, and conversation evidence (SKILL.md).\n
- Boundary markers: The skill uses a mandatory manual checkpoint (CHECKPOINT STOP) requiring explicit user approval before any file modifications, though it lacks automated delimiters for external content.\n
- Capability inventory: File system read/write access to CONTEXT.md and the docs/adr/ directory.\n
- Sanitization: No programmatic sanitization or filtering of external content is specified before processing.
Audit Metadata