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 reference guide for developers building MCP servers. It contains educational content, architectural patterns, and code snippets following industry best practices.
  • [SAFE]: The security section proactively identifies and provides mitigations for common AI agent vulnerabilities, such as command injection (citing CVE-2025-53967), tool poisoning, and supply chain risks.
  • [SAFE]: All code examples demonstrate secure usage of the Model Context Protocol SDK, specifically recommending per-request server instances to prevent cross-client data leakage (referencing GHSA-345p-7cg4-v4c7).
  • [SAFE]: Dependencies referenced are official packages from the Model Context Protocol organization and well-known web frameworks like Hono and Express.
  • [SAFE]: The skill correctly identifies and warns against dangerous practices like interpolating user input into shell commands or using shared server instances in stateless environments.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 12:06 AM