My MCP Server

🚀 My MCP Server

My MCP server enables applications to provide context to LLMs through the MCP protocol, enhancing AI interaction capabilities.

🚀 Quick Start

Step 1: Download Claude Desktop

Visit Claude Desktop Download to get the application.

Step 2: Build a Custom MCP Server

Execute the following commands in your terminal:

$ npm install
$ npm run build

# Display the full path of build/main.js
$ find `pwd` -maxdepth 2 -name main.js

Step 3: Register the MCP Server in Claude Desktop Settings

⚠️ Important Note

The following steps are for the macOS version. For other operating systems, please refer to the official documentation.

Open the configuration file:

$ vi ~/Library/Application\ 支持/Claude/claude_desktop_config.json

Modify the configuration file as follows:

{
  "mcpServers": {
    "myMcp": {
      // Note: Specify the full path of node (`$ which node`)
      "command": "node",
      // Note: Specify the full path of build/main.js
      "args": ["xxx/my-mcp-server/build/main.js"]
    }
  }
}

Step 4: Confirm the MCP Server is Registered in Claude Desktop

Restart Claude Desktop and confirm that the MCP server has been successfully registered.

Try asking about the weather in San Francisco. You'll see that the MCP server fetches the weather forecast information and integrates it into the AI-generated message.

(Reference) What it looks like without an MCP server:

✨ Features

MCP (Model Context Protocol)

MCP is a standardized protocol that allows applications to provide context to LLMs. It's like the USB-C port for AI applications, enabling a unified way to connect AI models with various data sources and tools.

flowchart LR
  User(fa:fa-user User)
  Client["""
  Host with MCP Client
  (Claude, IDEs, Tools)
  """]

  Server1["MCP Server A"]
  Server2["MCP Server B"]
  Server3["MCP Server C"]

  DB1[Local Data Source A]
  DB2[Local Data Source B]
  Services[Remote Services]

  subgraph [Local Computer]
  User ↔ Client
  Client ↔ |MCP Protocol| Server1
  Client ↔ |MCP Protocol| Server2
  Client ↔ |MCP Protocol| Server3

  Server1 ↔ DB1
  Server2 ↔ DB2
  end

  subgraph [Internet]
  Server2 ↔ |Network API| Services
  Server3 ↔ |Network API| Services
  end

Reference: MCP Introduction

Similar Concept: Function Calling

flowchart LR
  User(fa:fa-user User)
  Server[Application Server]
  OpenAI[OpenAI API]

  Function1[Function A on Server]
  Function2[Function B on Server]
  Function3[Function C on Server]
  DB[Local Data Source]
  Services[Remote Services]

  subgraph [Local Computer]
  User ↔ Server
  Server ↔ Function1
  Server ↔ Function2
  Server ↔ Function3
  Function1 ↔ DB
  Function2 ↔ DB
  end

  subgraph [Internet (OpenAI)]
  Server → |Prompt| OpenAI
  OpenAI → |Response with Function Call| Server
  end

  subgraph [Internet]
  Function2 ↔ |Network API| Services
  Function3 ↔ |Network API| Services
  end

Reference: OpenAI Function Calling Guide

📚 Documentation

Reference Links

• Anthropic MCP Announcement
• MCP Server Quickstart