MCP Server Template

MCP Server Template

A starter template for creating Model Context Protocol servers that integrate with Claude Desktop and other MCP-compatible clients. Provides a configurable tool system, Docker support, and TypeScript-ready structure for building custom MCP servers.

Category
访问服务器

README

MCP Server Template

A starter template for creating Model Context Protocol (MCP) servers that integrate with Claude Desktop and other MCP-compatible clients.

For details, check Anthroic mcp documentation https://docs.anthropic.com/en/docs/mcp

Features

  • Stdio transport implementation, including Claude Desktop integration
  • Configurable tool system with built-in and custom tools
  • JSON-based configuration with environment variable overrides
  • Docker support for containerized deployment
  • Comprehensive logging and error handling
  • TypeScript-ready structure
  • Example tools for common operations

Quick Start

Installation

# Clone the template
git clone https://github.com/yourusername/mcp-server-template.git
cd mcp-server-template

# Install dependencies
npm install

# Run in development mode
npm run dev

Configuration

Create a config.json file in the project root:

{
  "serverName": "my-mcp-server",
  "version": "1.0.0",
  "debug": true,
  "dataDir": "./data",
  "outputDir": "./output",
  "tools": {
    "enabled": true,
    "customToolsDir": "./custom-tools"
  }
}

Claude Desktop Integration

Add to your Claude Desktop configuration (claude_desktop_config.json):

{
  "mcpServers": {
    "my-server": {
      "command": "node",
      "args": ["/path/to/mcp-server-template/src/server/index.js"]
    }
  }
}

Project Structure

mcp-server-template/
├── src/
│   ├── server/          # Core server implementation
│   │   └── index.js     # Main server entry point
│   ├── config/          # Configuration management
│   │   └── config-loader.js
│   ├── tools/           # Tool system
│   │   └── tool-registry.js
│   └── generator/       # Template generator (optional)
├── examples/            # Example implementations
│   ├── custom-tool.js   # Example custom tool
│   └── docker/          # Docker configuration
├── tests/               # Test suite
├── docs/                # Documentation
├── config.json          # Default configuration
└── package.json

Creating Custom Tools

Create a new tool in the custom-tools directory:

// custom-tools/my-tool.js
export default {
  name: 'my_custom_tool',
  description: 'Description of what this tool does',
  inputSchema: {
    type: 'object',
    properties: {
      param1: {
        type: 'string',
        description: 'First parameter'
      },
      param2: {
        type: 'number',
        description: 'Second parameter'
      }
    },
    required: ['param1']
  },
  handler: async (args) => {
    // Tool implementation
    const result = await doSomething(args.param1, args.param2);
    return {
      success: true,
      result: result
    };
  }
};

Built-in Tools

The template includes several built-in tools:

  • list_items - List all items with optional filtering
  • add_item - Add a new item to the system
  • remove_item - Remove an item by ID
  • search_items - Search items with advanced queries
  • get_status - Get system status and statistics

Environment Variables

Override configuration using environment variables:

  • MCP_SERVER_NAME - Server name
  • MCP_VERSION - Server version
  • MCP_DEBUG - Enable debug logging
  • MCP_DATA_DIR - Data directory path
  • MCP_OUTPUT_DIR - Output directory path
  • MCP_CONFIG_PATH - Custom config file path

Docker Deployment

Build and run with Docker:

# Build image
docker build -t my-mcp-server .

# Run container
docker run -d \
  -v $(pwd)/data:/app/data \
  -v $(pwd)/output:/app/output \
  --name my-mcp-server \
  my-mcp-server

Development

Running Tests

npm test

Linting

npm run lint

Building for Production

npm run build

API Reference

Tool Schema

All tools follow this schema:

{
  name: string,           // Unique tool identifier
  description: string,    // Human-readable description
  inputSchema: {         // JSON Schema for parameters
    type: 'object',
    properties: {},
    required: []
  },
  handler: async function // Tool implementation
}

Configuration Schema

{
  serverName: string,     // Server identifier
  version: string,        // Server version
  debug: boolean,         // Enable debug logging
  dataDir: string,        // Data storage directory
  outputDir: string,      // Output directory
  maxFileSize: number,    // Maximum file size in bytes
  timeout: number,        // Request timeout in ms
  tools: {
    enabled: boolean,     // Enable tool system
    customToolsDir: string // Custom tools directory
  }
}

Extending the Template

Adding New Features

  1. Create feature modules in src/features/
  2. Register with the server in src/server/index.js
  3. Add configuration options to config-loader.js
  4. Document in README and API reference

Creating Domain-Specific Servers

  1. Fork this template
  2. Replace example tools with domain-specific tools
  3. Customize configuration schema
  4. Update documentation
  5. Deploy as standalone MCP server

Troubleshooting

Common Issues

Server not starting:

  • Check Node.js version (>=16.0.0)
  • Verify configuration file syntax
  • Check file permissions

Tools not loading:

  • Verify tool file exports default object
  • Check tool schema validity
  • Review debug logs for errors

Claude Desktop integration:

  • Ensure correct path in configuration
  • Check server process is running
  • Review Claude Desktop logs

Contributing

Contributions are welcome! Please:

  1. Fork the repository
  2. Create a feature branch
  3. Add tests for new features
  4. Update documentation
  5. Submit a pull request

License

MIT License - see LICENSE file for details

Support

Acknowledgments

Built for the Model Context Protocol ecosystem, enabling AI assistants to interact with external systems through a standardized interface.

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