mcp-best-practices

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a comprehensive developer reference for the Model Context Protocol. It includes detailed guidance on transport selection, tool design, and error handling for production environments.
  • [SAFE]: The documentation actively promotes security best practices, explicitly identifying top threats like command injection (CVE-2025-53967) and supply chain risks, and providing mitigation strategies such as input validation and DNS rebinding protection.
  • [SAFE]: Code examples provided are architectural templates using standard, well-known libraries (e.g., Zod, Hono, Express, and the official @modelcontextprotocol SDKs).
  • [SAFE]: No obfuscation, prompt injection, credential harvesting, or unauthorized remote code execution patterns were detected in the skill files.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 12:06 AM