Minimal MCP Server

Minimal MCP Server

A minimal MCP server boilerplate in Python for testing client-server communication, providing a simple 'hello' tool to verify connection.

Category
访问服务器

README

Minimal MCP Server Boilerplate

A minimal Model Context Protocol (MCP) server implementation in Python for testing client-server communication.

MCP Server

What This Does

This boilerplate provides the absolute minimum code needed to:

  • ✅ Connect with MCP clients (Cursor, Claude Desktop, etc.)
  • ✅ Respond to initialization requests
  • ✅ List available tools
  • ✅ Execute a simple "hello" tool

Quick Start

1. Prerequisites

  • Python 3.9 or higher
  • No external dependencies required!

2. Test Locally

# Make server executable
chmod +x server.py

# Run the test client
python test_client.py

You should see:

============================================================
Testing MCP Server
============================================================

1. Testing initialize...
✓ Initialize successful
  Server: minimal-mcp-server

2. Testing tools/list...
✓ Found 1 tool(s):
  - hello: A simple hello tool to test MCP connection

3. Testing tools/call (hello)...
✓ Tool call successful:
  Hello, Developer! MCP connection is working! 🎉

============================================================
All tests passed! 🎉
============================================================

Connecting to MCP Clients

Cursor IDE

  1. Find your Cursor MCP config file:

    • macOS: ~/Library/Application Support/Cursor/User/globalStorage/rooveterinaryinc.roo-cline/settings/cline_mcp_settings.json
    • Windows: %APPDATA%\Cursor\User\globalStorage\rooveterinaryinc.roo-cline\settings\cline_mcp_settings.json
    • Linux: ~/.config/Cursor/User/globalStorage/rooveterinaryinc.roo-cline/settings/cline_mcp_settings.json
  2. Add this configuration (update paths to match your system):

    {
      "mcpServers": {
        "minimal-mcp-server": {
          "command": "python3",
          "args": ["/absolute/path/to/mcp-boilerplate/server.py"],
          "cwd": "/absolute/path/to/mcp-boilerplate"
        }
      }
    }
    
  3. Restart Cursor and verify connection in settings

  4. Test: Ask Cursor to "Call the hello tool with name 'YourName'"

Claude Desktop

  1. Find Claude's config file:

    • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
    • Windows: %APPDATA%\Claude\claude_desktop_config.json
  2. Add the same configuration as above

  3. Restart Claude Desktop

  4. Test: Ask Claude to use the hello tool

Project Structure

mcp-boilerplate/
├── server.py          # Main MCP server implementation
├── test_client.py     # Local testing script
├── requirements.txt   # Dependencies (empty - uses stdlib)
├── mcp-config.json    # Example client configuration
└── README.md          # This file

How It Works

MCP Protocol Flow

  1. Initialize: Client sends initialization request

    {"method": "initialize", "params": {...}}
    
  2. List Tools: Client requests available tools

    {"method": "tools/list"}
    
  3. Call Tool: Client executes a tool

    {"method": "tools/call", "params": {"name": "hello", "arguments": {...}}}
    

Key Components

  • MinimalMCPServer: Main server class
  • get_tools(): Returns tool definitions
  • call_tool(): Executes tool logic
  • handle_request(): Routes JSON-RPC requests
  • run_stdio(): Main loop reading from stdin, writing to stdout

Extending the Boilerplate

Adding a New Tool

  1. Add tool definition in get_tools():

    {
        "name": "my_tool",
        "description": "Does something useful",
        "inputSchema": {
            "type": "object",
            "properties": {
                "param": {"type": "string", "description": "..."}
            },
            "required": ["param"]
        }
    }
    
  2. Add handler in call_tool():

    if name == "my_tool":
        result = do_something(arguments.get("param"))
        return {"type": "text", "text": result}
    

Adding Dependencies

If you need external libraries:

  1. Add to requirements.txt:

    requests>=2.31.0
    
  2. Install:

    pip install -r requirements.txt
    

Troubleshooting

Server not connecting?

  • Verify Python path is correct in config
  • Use absolute paths (not relative)
  • Check file permissions (chmod +x server.py)

"Method not found" errors?

  • Ensure method names match exactly (case-sensitive)
  • Check JSON-RPC format is correct

Server crashes?

  • Verify Python 3.9+ is installed
  • Check for syntax errors: python -m py_compile server.py

Next Steps

Once connection is verified:

  • Add your custom tools
  • Integrate with databases, APIs, or services
  • Add error handling and logging
  • Implement MCP resources and prompts

Resources

License

MIT License - feel free to use this boilerplate for your projects!

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选