MCP Integration Assistant

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides instructions and templates for building MCP servers using official libraries.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill explicitly instructs the agent to handle credentials securely using environment variables and warns against hardcoding secrets.
  • [EXTERNAL_DOWNLOADS]: References to @modelcontextprotocol/sdk and @modelcontextprotocol/inspector are to official, well-known packages in the MCP ecosystem.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 05:57 AM
Security Audit — agent-trust-hub — MCP Integration Assistant