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 process covering all of it. ## Core Features & Use Cases - End-to-End MCP Development Workflow: Walks through research, implementation, review, and evaluation phases for Python (FastMCP) or TypeScript (MCP SDK) servers. - Language-Specific Reference Guides: Includes detailed implementation patterns for Zod schemas, tool registration, pagination, error handling, and transport configuration (stdio and Streamable HTTP). - Evaluation Harness: Ships Python scripts to create XML-based QA evaluations and run them against your server using Claude, measuring accuracy and tool-call behavior. - Use Case: You want to expose a GitHub Projects API to an AI agent. Follow the guide to design well-named tools with proper annotations, implement them in TypeScript, then generate 10 complex read-only evaluation questions and run the harness to verify LLMs can answer them. ## Quick Start Ask the agent to help you design and implement an MCP server for your chosen API, following the four-phase workflow in this Skill.
