What problem does it solve? Building a Model Context Protocol server that LLMs can actually use effectively requires careful tool design, schema validation, transport selection, and evaluation—this Skill provides a structured four-phase workflow covering research, implementation, review, and evaluation. ## Core Features & Use Cases - Guided Four-Phase Workflow: Walks through deep research of the MCP specification and SDKs, implementation with Zod or Pydantic schemas, code review and testing with MCP Inspector, and evaluation creation. - Language-Specific References: Includes detailed implementation guides for TypeScript (registerTool, Zod schemas, streamable HTTP/stdio transports) and Python (FastMCP patterns), plus MCP best practices for naming, pagination, and error handling. - Evaluation Harness: Ships Python scripts to run XML-based QA evaluations against a live MCP server over stdio, SSE, or HTTP, producing accuracy reports with per-task tool-call metrics. - Use Case: You want to expose the GitHub API to an LLM agent. Use this Skill to plan tool coverage, implement a TypeScript MCP server with validated inputs and structured outputs, then generate 10 read-only evaluation questions and measure how well an agent answers them. ## Quick Start Ask the agent to design and build an MCP server for your chosen API using the mcp-builder workflow, starting with research and ending with an evaluation run.
