Mcp Learn
What is Mcp Learn
mcp_learn is a simple implementation example of the Model Context Protocol (MCP), which includes an MCP server and client for personal learning purposes.
Use cases
Use cases include building a personal assistant, querying weather information, performing calculations, and connecting multiple clients to a single server instance.
How to use
To use mcp_learn, copy the example configuration file and fill in your API key. You can run the client in two modes: standard input/output mode or Server-Sent Events (SSE) mode.
Key features
Key features include support for multiple tool calls (simulated), chain tool calls, configuration management (environment variables and config files), and multiple connection modes (stdio and SSE).
Where to use
mcp_learn can be used in various fields such as educational purposes, prototyping applications that require interaction with large language models, and testing tool integrations.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Overview
What is Mcp Learn
mcp_learn is a simple implementation example of the Model Context Protocol (MCP), which includes an MCP server and client for personal learning purposes.
Use cases
Use cases include building a personal assistant, querying weather information, performing calculations, and connecting multiple clients to a single server instance.
How to use
To use mcp_learn, copy the example configuration file and fill in your API key. You can run the client in two modes: standard input/output mode or Server-Sent Events (SSE) mode.
Key features
Key features include support for multiple tool calls (simulated), chain tool calls, configuration management (environment variables and config files), and multiple connection modes (stdio and SSE).
Where to use
mcp_learn can be used in various fields such as educational purposes, prototyping applications that require interaction with large language models, and testing tool integrations.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Content
MCP 客户端与服务器示例（个人学习）
这是一个基于 Model Context Protocol (MCP) 的简单实现示例，包含了一个 MCP 服务器和客户端。
项目结构
testsever/main.py: MCP 服务器实现，测试工具功能mcp_client.py: MCP 客户端实现，连接到服务器并处理用户查询config.py: 配置管理model_config.py: 模型配置管理config.json.example: 配置文件示例，实际重命名为config.json
功能特性
- 支持多种工具调用（计算器、天气查询、网页获取等（功能是模拟的））
- 支持链式工具调用
- 配置管理（支持环境变量和配置文件）
- 支持多种连接模式：
- 标准输入输出(stdio)模式：直接启动服务器并连接
- 服务器发送事件(SSE)模式：通过HTTP连接到运行中的服务器
使用方法
- 复制配置文件示例并填写你的 API 密钥：
cp config.json.example config.json
- 编辑
config.json文件，填入你的阿里云千问 API 密钥/也可以是其他openai格式的支持LLM的大模型api
标准输入输出(stdio)模式
直接运行客户端连接到服务器：
python mcp_client.py testsever/main.py
//或者启动某些支持mcp的库比如mcp_server_fetch
python mcp_client.py -m mcp_server_fetch
//或者npx启动(如果装了npx)，这种方法会自动识别目录下的.venv
cd <环境目录>
npx @modelcontextprotocol/inspector python main.py
//本质上是
npx @modelcontextprotocol/inspector -- "<python解释器路径>" "<服务器脚本路径>"
//举例
npx @modelcontextprotocol/inspector -- "E:/1awd/zm/pro3/mcp/testsever/.venv/Scripts/python.exe" "E:/1awd/zm/pro3/mcp/testsever/main.py"
这种模式下，客户端会自动启动服务器进程并通过标准输入输出进行通信。
服务器发送事件(SSE)模式
- 先启动SSE服务器：
python testsever/main.py --mode http --host 127.0.0.1 --port 8000
- 然后在另一个终端窗口中运行客户端连接到SSE服务器：
python mcp_client.py --mode sse http://127.0.0.1:8000/sse
这种模式下，服务器作为一个HTTP服务运行，客户端通过SSE协议连接到它。这允许多个客户端连接到同一个服务器实例。
运行效果
环境变量
你也可以通过环境变量来配置：
ALIYUN_API_KEY: 阿里云千问 API 密钥ALIYUN_API_BASE: API 基础 URLALIYUN_MODEL: 使用的模型名称
环境配置
- 参考官网
MCP协议说明
MCP（Model Context Protocol）是一个用于大型语言模型与工具之间交互的协议。本项目实现了两种MCP连接方式：
- stdio模式：通过标准输入输出进行通信，适合单用户场景
- SSE模式：通过HTTP的Server-Sent Events进行通信，支持多用户连接
两种模式都符合官方MCP标准，支持相同的工具调用功能。
参考资料
Dev Tools Supporting MCP
The following are the main code editors that support the Model Context Protocol. Click the link to visit the official website for more information.
