guard
Pass
Audited by Gen Agent Trust Hub on Apr 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a procedural guide for implementing security 'guards' (input/output validation, cost controls, and permission scoping). It contains no executable scripts, shell commands, or network operations.
- [SAFE]: The content promotes defensive engineering best practices, such as sanitizing for injection patterns, PII masking, and requiring confirmation for destructive operations.
- [SAFE]: No obfuscation, prompt injection attempts, credential exposure, or data exfiltration patterns were detected in the instructions or metadata.
Audit Metadata