adversarial-self-review
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill provides natural language guidelines for adversarial self-review and falsification of conclusions. These are benign instructions for improving response quality.
- [NO_CODE]: The skill does not include any scripts, command executions, or external package requirements.
Audit Metadata