AI Guardian MCP Server

AI Guardian is a well-architected security governance server for local LLMs with comprehensive audit logging, policy controls, and input validation. The codebase demonstrates strong security awareness and follows defense-in-depth principles. However, several findings warrant attention: (1) environment variable-based secret passing for Ollama authentication is inherently risky despite being optional, (2) the governance decorator logs parameter names which could indirectly leak secrets in certain edge cases, and (3) the pattern engine's file-based state is not persistent across restarts, limiting effectiveness of circuit breakers. These are moderate concerns appropriate to the preview/v0.1 status but should be addressed before production use. Supply chain analysis found 5 known vulnerabilities in dependencies (0 critical, 2 high severity). Package verification found 1 issue.

5 files analyzed · 14 issues found

Security scores are indicators to help you make informed decisions, not guarantees. Always review permissions before connecting any MCP server.