MCP Integration

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is entirely composed of instructional content and configuration templates designed to teach the agent how to handle MCP server integrations.
  • [SAFE]: The documentation explicitly promotes security best practices, including the use of environment variables for secrets (CREDENTIALS_SAFE), the requirement for secure protocols (HTTPS/WSS), and the principle of least privilege for tool permissions.
  • [SAFE]: All identified external links point to official documentation (Claude, Model Context Protocol) or well-known, trusted platforms (Asana, GitHub).
  • [SAFE]: Example code snippets for server configuration and authentication scripts are benign templates intended for developer customization and do not contain hidden malicious logic or obfuscation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 04:52 PM
Security Audit — agent-trust-hub — MCP Integration