eval-harness
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill provides conceptual instructions and scoring rubrics for benchmarking AI agents but includes no script files (Python, JavaScript, Shell) or automated execution logic.
- [SAFE]: No malicious patterns, obfuscation, or data exfiltration vectors were identified in the markdown content or YAML frontmatter.
Audit Metadata