llm-wiki-security-review
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a defensive auditing tool that guides agents in identifying security risks in other systems. It does not contain any code or instructions that would allow for prompt injection, data exfiltration, or unauthorized command execution.
- [SAFE]: All external URLs point to reputable and trusted sources, such as the NIST National Vulnerability Database (NVD), arXiv for security research papers, and official documentation for well-known security tools like Gitleaks, Semgrep, and Microsoft Presidio.
- [SAFE]: The skill explicitly incorporates safety gates, such as requiring human review for sensitive writes, mandating the use of sandboxes for document parsing, and defining clear protocols for handling sensitive or regulated data.
Audit Metadata