risk-calibrated-pr-review
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows security best practices for AI agents. It explicitly mandates a read-only posture, forbidding file edits, metadata changes, or the execution of external operations. It is specifically designed to handle untrusted input (Pull Requests) by instructing the agent to assume all implementation-authored evidence is incorrect until independently verified against an authoritative source. This defensive posture effectively mitigates common indirect prompt injection and deception risks inherent in code review tasks.
Audit Metadata