ReviewAgent

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a template for a validation agent that processes structured JSON input to provide a review verdict. It contains no executable code or tool definitions.
  • [SAFE]: There are no indicators of data exfiltration, as the skill does not request network permissions or include URL references to external domains.
  • [SAFE]: No prompt injection or safety bypass patterns were detected. The instructions clearly define a role and constraints for the agent to follow for its intended validation purpose.
  • [SAFE]: The skill does not access sensitive system files or credentials. It operates entirely on the input provided in the schema.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 06:34 AM