plan-review
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external plan files, creating an injection surface.
- Ingestion points: The agent reads user-specified plan files and review logs as part of its core workflow.
- Boundary markers: The prompt template used to invoke the external reviewer does not employ delimiters to isolate the plan data from instructions.
- Capability inventory: The agent has permissions to read and write files and invoke secondary agents, allowing injected content to influence the file system.
- Sanitization: There is no evidence of sanitization or safety filtering for the input plan content.
Audit Metadata