Which programming languages does this skill support?

This skill provides specialized implementation guidance for both TypeScript (using the official MCP SDK) and Python (using the FastMCP framework).

How do I test my MCP server implementation?

You can use the MCP Inspector for local debugging or follow the skill's evaluation phase to create XML-based QA pairs for automated performance testing.

What is an MCP server?

An MCP (Model Context Protocol) server is a standardized interface that allows Large Language Models to connect securely with external data sources, tools, and services.

What is the difference between stdio and HTTP transport in MCP?

stdio is designed for local servers running on the same machine as the AI client, while streamable HTTP is preferred for remote, scalable server deployments.

MCP Server Builder

Name: MCP Server Builder
Author: bmad-labs

bybmad-labs

•

API 开发

Builds high-quality Model Context Protocol (MCP) servers to bridge LLMs with external APIs and services using Python or TypeScript.

The mcp-builder skill provides a comprehensive, phased workflow for designing, implementing, and testing Model Context Protocol (MCP) servers. It guides developers through deep research into API coverage, project structure setup for both Python (FastMCP) and TypeScript (MCP SDK), and the implementation of well-documented tools with precise schemas. By following industry best practices for tool naming, context management, and error handling, this skill ensures that external services are seamlessly and effectively accessible to AI models through standardized interfaces.

主要功能

01Support for Zod and Pydantic schema definitions for robust input/output validation

02Structured workflow for TypeScript (MCP SDK) and Python (FastMCP) server development

03Automated evaluation generation to verify LLM performance using XML QA pairs

04Guidelines for creating clear tool names and discoverable API endpoints

051 GitHub stars

06Reference library for MCP protocol specifications and transport mechanisms

使用场景

01Developing and testing robust evaluations to ensure an AI agent can reliably execute complex tool-calling sequences

02Creating custom MCP servers to allow Claude to interact with internal company databases or proprietary APIs

03Building standardized integrations for third-party SaaS platforms like GitHub, Slack, or Jira

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add bmad-labs/skills mcp-builder

For use in Claude.ai and ChatGPT

Download Skill