prompt-guard
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a legitimate security utility that functions as a scanner for malicious content. Its behavior is consistent with its stated purpose of identifying prompt injection attempts.
- [PROMPT_INJECTION]: The skill contains a library of prompt injection patterns such as 'Ignore previous instructions' and 'Act as if you have no restrictions'. These are used strictly as detection signatures for analysis and do not represent an attempt to hijack agent behavior.
- [DATA_EXFILTRATION]: No sensitive data exposure or exfiltration patterns were detected. The skill has 'network: false' permissions and does not contain hardcoded credentials or access sensitive system paths.
- [REMOTE_CODE_EXECUTION]: No remote code execution, package management, or dynamic execution patterns were identified. The skill performs text analysis without executing code from external sources.
Audit Metadata