build-mcp-sdk-v2

Pass

Audited by Gen Agent Trust Hub on Apr 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a development guide for the Model Context Protocol (MCP) v2 SDK. It correctly identifies and promotes the use of official, split-package libraries such as @modelcontextprotocol/server, @modelcontextprotocol/client, and @modelcontextprotocol/core.
  • [SAFE]: Implementation examples demonstrate security-conscious patterns, including the use of environment variables for sensitive credentials (API keys, tokens) and the application of DNS rebinding protection through official framework adapters for Express and Hono.
  • [SAFE]: Tool definitions use Zod v4 for strict input and output schema validation, which helps mitigate potential injection and data integrity issues when processing external data.
  • [SAFE]: No instances of prompt injection, data exfiltration, obfuscation, or unauthorized command execution were found. All development commands (e.g., npm install, npx @anthropic-ai/mcp-inspector) are standard and appropriate for the skill's documented purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 10, 2026, 12:51 PM