agent-evaluation
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill provides conceptual frameworks and best practices for agent evaluation. It does not contain any executable scripts, shell commands, or network requests.
- [NO_CODE]: There are no implementation files or scripts associated with this skill, precluding risks related to remote code execution, persistence, or privilege escalation.
- [METADATA_POISONING]: Metadata fields (name, description, source) are consistent with the content and do not contain deceptive or malicious instructions.
Audit Metadata