exemplar-review
Pass
Audited by Gen Agent Trust Hub on Aug 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a purely instructional reviewer designed to provide qualitative feedback on code diffs based on established industry standards (prior-art and excellence-bars).
- [SAFE]: The skill incorporates safety constraints for automated environments, specifically instructing the agent to remain in a 'read-only' mode when invoked within orchestrators (e.g.,
/review-fix), explicitly forbidding unauthorized editing, committing, or pushing of code. - [SAFE]: The skill references and routes to highly reputable and trusted sources for its evaluation criteria, including documentation from Anthropic, Stripe, Apple, and the Python core team.
- [SAFE]: The skill instructions emphasize 'Lens, not oracle', requiring the agent to ground all findings in citable, shipped work or documented principles, which reduces the risk of hallucinations or arbitrary behavior.
Audit Metadata