code-review-excellence
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes pull requests and code changes which are untrusted external inputs. This creates a surface for indirect prompt injection where malicious instructions embedded in the code could influence the agent's review behavior.
- Ingestion points: Pull request content and code changes referenced in SKILL.md.
- Boundary markers: None identified in the instructions to delimit external code from system instructions.
- Capability inventory: The skill generates text-based feedback; it lacks network or file-system write capabilities in the provided files.
- Sanitization: No sanitization or filtering of input code is defined.
Audit Metadata