opc-conversion-loop
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from previous project stages, creating a potential surface for indirect instructions to influence behavior.
- Ingestion points: Files located in
opc-doc/outputs/06-mvp-design/*are used as primary input. - Boundary markers: The prompt template lacks specific delimiters to separate user data from instructions.
- Capability inventory: The skill has permissions to write to the
opc-doc/directory, including updating stage and decision state files. - Sanitization: No explicit input validation or sanitization routines are defined for the ingested file content.
Audit Metadata