composite-scorer
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, prompt injections, or unauthorized network/file access instructions were detected. The skill is primarily focused on logic and formatting for assessment purposes.
- [SAFE]: The skill implements a rigid JSON output contract for its scoring results. This structured approach is a security best practice that helps prevent indirect prompt injection by ensuring the agent's output remains within a predictable schema regardless of the content being analyzed.
Audit Metadata