file-links
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a robust framework for identifying security risks in AI agent extensions without implementing any malicious operations itself.
- [SAFE]: Examples of vulnerabilities such as prompt injection and data exfiltration are included solely for pattern-matching guidance.
- [SAFE]: No obfuscation, unauthorized command execution, or persistence mechanisms were found in the provided documentation.
Audit Metadata