Skip to main content
Glama

PyMCP Sum Server

by ChaosXu

pymcp

a MCP server

SDK

https://github.com/modelcontextprotocol/python-sdk

项目结构

src/ └── mcp_server/ ├── __init__.py └── sum_int.py # MCP服务器实现,提供两个整数相加功能 tests/ ├── test_sum_int.py # 基础功能测试 ├── test_sum_int_with_real_llm.py # 真实LLM调用测试 ├── test_sum_int_with_agent.py # 使用LangChain Agent的测试 (stdio方式) ├── test_sum_int_with_agent_sse.py # 使用LangChain Agent的测试 (SSE方式) └── test_sum_int_with_agent_streamable_http.py # 使用LangChain Agent的测试 (Streamable HTTP方式)

环境初始化

安装依赖

使用 uv 管理依赖:

# 安装生产依赖 uv pip install -e . # 安装测试依赖(包括生产依赖) uv pip install -e .[test]

环境变量配置

复制 .env.example 文件并重命名为 .env,然后根据你的实际情况修改其中的值:

cp .env.example .env

环境变量说明:

  • LLM_BASE_URL:LLM API的基础URL
  • LLM_API_KEY:访问LLM API的密钥
  • LLM_MODEL:要使用的模型名称
  • MCP_SERVER_PORT:MCP服务器端口(默认为8000)

运行MCP服务器

MCP服务器支持多种传输方式:

stdio方式(默认)

# 运行MCP服务器(stdio方式) uv run mcp dev src/mcp_server/sum_int.py # 或者直接运行 python src/mcp_server/sum_int.py # 或者显式指定stdio方式 python src/mcp_server/sum_int.py stdio

SSE方式

# 使用SSE方式运行 python src/mcp_server/sum_int.py sse

服务器将在 http://127.0.0.1:8000 启动

Streamable HTTP方式

# 使用Streamable HTTP方式运行 python src/mcp_server/sum_int.py streamable-http

服务器将在 http://127.0.0.1:8000 启动

执行测试

基础功能测试

python tests/test_sum_int.py

真实LLM调用测试

python tests/test_sum_int_with_real_llm.py

使用LangChain Agent的测试 (stdio方式)

python tests/test_sum_int_with_agent.py

使用LangChain Agent的测试 (SSE方式)

python tests/test_sum_int_with_agent_sse.py

使用LangChain Agent的测试 (Streamable HTTP方式)

python tests/test_sum_int_with_agent_streamable_http.py

依赖说明

生产依赖:

  • mcp[cli]>=1.12.4 - MCP Python SDK

测试依赖:

  • openai>=1.99.9 - OpenAI Python客户端
  • python-dotenv>=1.0.1 - 环境变量加载工具
  • httpx[socks]>=0.28.1 - HTTP客户端(支持SOCKS代理)
  • langchain>=0.3.27 - LangChain核心库
  • langchain-openai>=0.3.29 - LangChain的OpenAI集成
  • langchain-mcp-adapters>=0.1.9 - LangChain与MCP的适配器
  • langgraph>=0.6.4 - LangGraph库,用于构建agent工作流
-
security - not tested
F
license - not found
-
quality - not tested

remote-capable server

The server can be hosted and run remotely because it primarily relies on remote services or has no dependency on the local environment.

A simple MCP server that provides integer addition functionality. Enables users to perform basic arithmetic operations by adding two integers together through natural language interactions.

  1. SDK
    1. 项目结构
      1. 环境初始化
        1. 安装依赖
        2. 环境变量配置
      2. 运行MCP服务器
        1. stdio方式(默认)
        2. SSE方式
        3. Streamable HTTP方式
      3. 执行测试
        1. 基础功能测试
        2. 真实LLM调用测试
        3. 使用LangChain Agent的测试 (stdio方式)
        4. 使用LangChain Agent的测试 (SSE方式)
        5. 使用LangChain Agent的测试 (Streamable HTTP方式)
      4. 依赖说明

        Related MCP Servers

        • A
          security
          F
          license
          A
          quality
          A simple and extendable MCP server that currently provides basic addition functionality and can be easily extended by defining new tools with docstrings.
          Last updated -
          4
          Python
        • -
          security
          F
          license
          -
          quality
          A server that provides basic mathematical operations (addition, subtraction, multiplication, division, power, square root) through MCP tools for use with AI assistants like Claude.
          Last updated -
          Python
        • -
          security
          F
          license
          -
          quality
          A containerized MCP (Model Context Protocol) server that provides a simple calculator tool for adding two numbers, deployable to Kubernetes/EKS environments.
          Last updated -
          4
          Python
        • -
          security
          F
          license
          -
          quality
          A simple calculator MCP server that provides basic arithmetic functions (add and subtract) through HTTP transport, designed to be cloud-ready and easily deployable.
          Last updated -
          Python
          • Linux
          • Apple

        View all related MCP servers

        MCP directory API

        We provide all the information about MCP servers via our MCP API.

        curl -X GET 'https://glama.ai/api/mcp/v1/servers/ChaosXu/pymcp'

        If you have feedback or need assistance with the MCP directory API, please join our Discord server