MCP Server Boilerplate

@cds-id

リポジトリ

基本情報

カテゴリ

開発者ツール

トランスポート

stdio

作者

cds-id

MCP Server Boilerplate について

NodeJS MCP Server Boilerplate

MCP Server Boilerplate の設定

標準の設定はありません

このサーバーの README には解析可能な MCP 設定ブロックが含まれていません。インストール手順はリポジトリをご確認ください。

リポジトリ

MCP Server Boilerplate のツール

`operation`: Operation to perform (add, subtract, multiply, divide)

`format`: Output format (iso, unix, readable)

概要

What is MCP Server Boilerplate?

MCP Server Boilerplate is a starter implementation of a server following the Model Context Protocol (MCP), built with TypeScript and Express. It provides sample resources, tools, and prompts to help developers quickly build MCP-compatible servers that supply context to LLMs in a standardized way.

How to use MCP Server Boilerplate?

Clone the repository, install dependencies (npm install), copy .env.example to .env, and configure environment variables like PORT. Run the HTTP server with npm start or npm run dev for development. For command-line tools, use npm run start:stdio or npm run dev:stdio.

Key features of MCP Server Boilerplate

Built with TypeScript and Express for HTTP transport
Includes sample resources, tools, and prompts
Supports both HTTP and stdio transport modes
Easy extension via modular resource/tool/prompt files
Uses the official MCP TypeScript SDK

Use cases of MCP Server Boilerplate

Rapidly prototype a new MCP server for LLM context
Learn how to implement resources, tools, and prompts with MCP
Serve as a foundation for production MCP servers
Test and debug MCP implementations using the Inspector tool

よくある質問

What are the prerequisites to run MCP Server Boilerplate?

Node.js v18 or later and npm or yarn are required.

How do I start the server?

Build the project with `npm run build`, then run `npm start` for HTTP mode or `npm run start:stdio` for stdio mode. For development with auto-restart, use `npm run dev` or `npm run dev:stdio`.

Can I add my own resources, tools, or prompts?

Yes. Create a new file in the respective `src/resources/`, `src/tools/`, or `src/prompts/` directory, implement the handler, and register it in the corresponding `index.ts` file. The README provides code examples for each.

What transport modes are supported?

HTTP (via Express on a configurable port) and stdio (for command-line integration) are both supported.

Is there a license for this project?

Yes, the project is licensed under the MIT License.

「開発者ツール」の他のコンテンツ

FARPY

mangomunchr

Verified distributed compute for humans, applications, and agents. Submit Blender, Octane, and video workloads, route them across qualified GPUs, track asynchronous execution, download finished artifacts, and verify FARP