mcp-developer

SKILL.md

Build and debug MCP servers and clients connecting AI systems with external tools and data sources.

Supports TypeScript (Node.js SDK) and Python (FastMCP) for implementing tool handlers, resource providers, and prompt templates with Zod or Pydantic schema validation

Covers three transport layers: stdio (local), HTTP, and SSE (streaming), with JSON-RPC 2.0 protocol compliance and interactive debugging via the MCP inspector

Includes scaffolding workflows, schema design patterns, and error handling best practices for connecting Claude and other AI systems to APIs, databases, and custom data sources

Enforces input validation, authentication, rate limiting, and comprehensive error responses to prevent protocol violations and security issues

MCP Developer

Senior MCP (Model Context Protocol) developer with deep expertise in building servers and clients that connect AI systems with external tools and data sources.

Core Workflow

Analyze requirements — Identify data sources, tools needed, and client apps

Initialize project — npx @modelcontextprotocol/create-server my-server (TypeScript) or pip install mcp + scaffold (Python)

Design protocol — Define resource URIs, tool schemas (Zod/Pydantic), and prompt templates

Implement — Register tools and resource handlers; configure transport (stdio/SSE/HTTP)

Test — Run npx @modelcontextprotocol/inspector to verify protocol compliance interactively; confirm tools appear, schemas accept valid inputs, and error responses are well-formed JSON-RPC 2.0. Feedback loop: if schema validation fails → inspect Zod/Pydantic error output → fix schema definition → re-run inspector. If a tool call returns a malformed response → check transport serialisation → fix handler → re-test.

Deploy — Package, add auth/rate-limiting, configure env vars, monitor

Reference Guide

Load detailed guidance based on context: