M

MCP Chinese Getting Started Guide

The Model Context Protocol (MCP) is an open - source protocol that provides a standardized method for large language models to connect to external data sources and tools. It supports functions such as resources, prompts, tools, and sampling. This article focuses on how to develop and use MCP tool services.
0 points
1.4K

What is an MCP server?

The MCP server is a standardized interface that allows large language models (LLMs) to connect to various data sources and tools, enabling seamless access and processing of information. It supports multiple functions such as resources, prompts, and tools.

How to use an MCP server?

You can quickly develop an MCP server using Python. It supports two transport protocols (stdio and SSE) and can be integrated into the Claude desktop client or other AI applications.

Use cases

It is suitable for scenarios where the functions of AI models need to be extended, such as web search, file operations, data queries, etc. It is particularly suitable for applications that require integrating AI with external systems.

Main Features

Resource ManagementProvides preset resources or custom protocol resources that can be directly called by AI models.
Prompt TemplatesCreates reusable prompt templates to simplify interactions with AI models.
Tool IntegrationSupports the development of custom tool functions to extend the capabilities of AI models.
Execution MonitoringProvides callback interfaces before and after tool execution for manual supervision or pre - processing.
Lifecycle ManagementSupports custom processing logic during service initialization, running, and shutdown.

Advantages and Limitations

Advantages
Standardized interface that unifies the interaction between different AI models and external systems
Easy to expand, supporting rapid development of new functions
Supports local and cloud deployment
Good tool development experience with debugging tools provided
Limitations
Requires some Python development knowledge
Some functions (such as resource templates) are not fully supported in the Claude client
stdio output is limited during debugging

How to Use

Install Dependencies
Use the uv tool to initialize the project and install the mcp package
Develop Tool Functions
Use the @tool decorator to create custom tools
Run the Server
Select the transport protocol (stdio or SSE) to start the service
Debug Tools
Use the MCP Inspector visualization tool for testing
Integrate into the Client
Configure the MCP server in Claude or other clients

Usage Examples

Web Search ToolImplement a tool that allows AI models to perform web searches
File Deletion ToolImplement a file deletion function that requires manual confirmation
Image Generation ToolIntegrate an AI image generation service into the Claude client

Frequently Asked Questions

How to debug an MCP server?
Why can't I see the resource templates in the Claude client?
How to deploy to the cloud?
Why can't I see the print output in stdio mode?

Related Resources

MCP Official Documentation
Official documentation for the MCP protocol
Python SDK Repository
Source code for the MCP Python SDK
Getting Started Guide Repository
Chinese getting - started guide and sample code
LangChain Adapter
Adapter for integrating MCP with LangChain
Installation
Copy the following command to your Client for configuration
{
  "mcpServers": {
    "web-search-server": {
      "command": "uv",
      "args": [
        "--directory",
        "D:/projects/mcp_getting_started",
        "run",
        "web_search.py"
      ]
    }
  }
}

{
  "mcpServers": {
    "image_server": {
      "command": "uv",
      "args": [
        "--directory",
        "D:/projects/mcp_getting_started",
        "run",
        "image_server.py"
      ],
      "env": {},
      "disabled": false,
      "autoApprove": []
    }
  }
}

{
  "mcpServers": {
    "web-search": {
      "url": "https://mcp-test-whhergsbso.cn-hangzhou.fcapp.run/sse"
    }
  }
}
Note: Your key is sensitive information, do not share it with anyone.
Z
Zen MCP Server
Zen MCP is a multi-model AI collaborative development server that provides enhanced workflow tools and cross-model context management for AI coding assistants such as Claude and Gemini CLI. It supports seamless collaboration of multiple AI models to complete development tasks such as code review, debugging, and refactoring, and can maintain the continuation of conversation context between different workflows.
Python
17
5 points
O
Opendia
OpenDia is an open - source browser extension tool that allows AI models to directly control the user's browser, perform automated operations using existing login status, bookmarks and other data, support multiple browsers and AI models, and focus on privacy protection.
JavaScript
14
5 points
C
Container Use
Container Use is an open-source tool that provides a containerized isolated environment for coding agents, supporting parallel development of multiple agents without interference.
Go
12
5 points
N
Notte Browser
Certified
Notte is an open-source full-stack network AI agent framework that provides browser sessions, automated LLM-driven agents, web page observation and operation, credential management, etc. It aims to transform the Internet into an agent-friendly environment and reduce the cognitive burden of LLMs by describing website structures in natural language.
701
4.5 points
S
Search1api
The Search1API MCP Server is a server based on the Model Context Protocol (MCP), providing search and crawling functions, and supporting multiple search services and tools.
TypeScript
375
4 points
D
Duckduckgo MCP Server
Certified
The DuckDuckGo Search MCP Server provides web search and content scraping services for LLMs such as Claude.
Python
890
4.3 points
B
Bing Search MCP
An MCP server for integrating Microsoft Bing Search API, supporting web page, news, and image search functions, providing network search capabilities for AI assistants.
Python
262
4 points
M
MCP Alchemy
Certified
MCP Alchemy is a tool that connects Claude Desktop to multiple databases, supporting SQL queries, database structure analysis, and data report generation.
Python
360
4.2 points
Featured MCP Services
N
Notion Api MCP
Certified
A Python-based MCP Server that provides advanced to-do list management and content organization functions through the Notion API, enabling seamless integration between AI models and Notion.
Python
199
4.5 points
M
Markdownify MCP
Markdownify is a multi-functional file conversion service that supports converting multiple formats such as PDFs, images, audio, and web page content into Markdown format.
TypeScript
1.8K
5 points
G
Gitlab MCP Server
Certified
The GitLab MCP server is a project based on the Model Context Protocol that provides a comprehensive toolset for interacting with GitLab accounts, including code review, merge request management, CI/CD configuration, and other functions.
TypeScript
152
4.3 points
D
Duckduckgo MCP Server
Certified
The DuckDuckGo Search MCP Server provides web search and content scraping services for LLMs such as Claude.
Python
890
4.3 points
F
Figma Context MCP
Framelink Figma MCP Server is a server that provides access to Figma design data for AI programming tools (such as Cursor). By simplifying the Figma API response, it helps AI more accurately achieve one - click conversion from design to code.
TypeScript
6.7K
4.5 points
U
Unity
Certified
UnityMCP is a Unity editor plugin that implements the Model Context Protocol (MCP), providing seamless integration between Unity and AI assistants, including real - time state monitoring, remote command execution, and log functions.
C#
614
5 points
M
Minimax MCP Server
The MiniMax Model Context Protocol (MCP) is an official server that supports interaction with powerful text-to-speech, video/image generation APIs, and is suitable for various client tools such as Claude Desktop and Cursor.
Python
796
4.8 points
G
Gmail MCP Server
A Gmail automatic authentication MCP server designed for Claude Desktop, supporting Gmail management through natural language interaction, including complete functions such as sending emails, label management, and batch operations.
TypeScript
333
4.5 points
AIbase
Zhiqi Future, Your AI Solution Think Tank
© 2025AIbase