adversarial-reviewer
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from git diffs and file reads, which creates an attack surface where instructions embedded in the code could attempt to influence the agent's review behavior.
- Ingestion points: The skill workflow gathers code changes via git diff commands and performs full file reads as described in Steps 1 and 2.
- Boundary markers: The instructions do not define specific delimiters or boundary markers to separate the review instructions from the code being analyzed.
- Capability inventory: The skill is restricted to reading local files and executing standard git commands to retrieve code differences.
- Sanitization: No sanitization or filtering of the ingested code content is performed before analysis by the LLM.
Audit Metadata