lean-review
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from the local repository during its audit process.
- Ingestion points: The skill reads uncommitted changes via
git diff(using thelean-review-enginetool) and accesses local project files such asspec.md,discuss.md, and.cocoplus/modes/lean.mode. - Boundary markers: The instructions do not define clear delimiters or specific instructions for the agent to treat content within the diff as data only, leaving an opening for instructions embedded in code or comments to influence the agent's behavior.
- Capability inventory: The skill invokes the
cocolean/lean-review-enginetool and performs filesystem checks for the existence of the.cocoplus/directory and specific mode files. - Sanitization: There is no evidence of sanitization or filtering applied to the content retrieved from the codebase before it is analyzed by the agent.
Audit Metadata