mcp-builder

Pass

Audited by Gen Agent Trust Hub on Jul 18, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch documentation and SDK guidelines from official Model Context Protocol sources (modelcontextprotocol.io) and their public GitHub repositories. These are recognized, well-known technology organizations and services.
  • [COMMAND_EXECUTION]: The evaluation script (scripts/evaluation.py) is designed to run local MCP servers for testing purposes via standard I/O. This is a standard development workflow for testing local code and is used within the intended scope of the skill's purpose as a developer tool.
  • [CREDENTIALS_UNSAFE]: The skill demonstrates safe credential management practices. The evaluation harness and documentation explicitly instruct users to provide API keys via environment variables or command-line arguments rather than hardcoding them.
  • [DATA_EXFILTRATION]: Network operations are limited to communicating with the Anthropic API (for the evaluation harness) and fetching documentation from trusted sources. No patterns of sensitive data exfiltration were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 18, 2026, 04:24 PM
Security Audit — agent-trust-hub — mcp-builder