mcp-server-security
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a security best-practice guide and does not contain any executable malicious code or instructions. It provides templates for defensive configurations such as TLS termination, OAuth 2.1 integration, and mTLS cert generation.
- [SAFE]: The provided code snippets demonstrate secure coding practices, including JSON Schema validation for tool inputs, parameterized SQL queries to prevent injection, and filesystem sandboxing to prevent path traversal.
- [SAFE]: The skill explicitly addresses security threats such as SSRF and prompt injection by providing mitigation strategies, including domain allowlisting and robust input sanitization guides.
- [SAFE]: Deployment hardening examples (Docker and Kubernetes) advocate for running as non-root users, using read-only filesystems, and applying seccomp profiles to minimize the attack surface.
Audit Metadata