agent-evaluation

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill is a purely instructional guide for developing evaluation frameworks for AI agents. It describes methodologies for case selection, grading techniques, and result reporting.
  • [NO_CODE]: There are no scripts, configuration files, shell commands, or network operations included in the skill. It poses no risk of code execution or data exposure.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 03:40 PM
Security Audit — agent-trust-hub — agent-evaluation