pretotype
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it is designed to read and process content from external files.
- Ingestion points: The skill reads content from user-provided file paths and specific directories such as
{Product}/Product/pretotypes/,{Product}/Product/experiments/, andcontext/documents/index.md. - Boundary markers: The instructions for the 'UPDATE' and 'FIND' modes do not define specific delimiters or safety warnings (e.g., 'ignore embedded instructions') for the content being ingested from these files.
- Capability inventory: The skill is capable of performing file writes (saving pretotype plans) and triggering other internal commands such as
/context-save,/compound, and/assumption-map. - Sanitization: There is no evidence of validation or sanitization of the content extracted from external documents before it is processed by the agent.
Audit Metadata