Claude Arithmetic MCP Server
An MCP server that exposes basic arithmetic operations (addition, subtraction, multiplication, division) as tools, usable locally with Claude Desktop or deployed remotely.
README
Claude Arithmetic MCP Server
A simple Model Context Protocol (MCP) project built with Python andFastMCP. The server exposes basic arithmetic operations as MCP tools andcan be used locally with Claude Desktop or deployed as a remote MCPserver.
Project Overview
This project demonstrates the basic MCP client-server architecture:
Claude Desktop | | MCP v Arithmetic MCP Server | +-- add() +-- subtract() +-- multiply() +-- divide()
The server was tested using MCP Inspector and deployed remotely usingPrefect Horizon.
Features
Addition
Subtraction
Multiplication
Division
MCP tool registration using FastMCP
Local testing with MCP Inspector
Claude Desktop integration
Remote deployment through Prefect Horizon
Tech Stack
Python
MCP
FastMCP
MCP Inspector
Claude Desktop
Prefect Horizon
Git & GitHub
MCP Tools
Tool Description
add Adds two numberssubtract Subtracts the second number from the firstmultiply Multiplies two numbersdivide Divides the first number by the second
The divide tool also handles division by zero.
Project Structure
MCP-Client-Server-Integration/ │ ├── server.py ├── fastmcp.json ├── pyproject.toml ├── uv.lock ├── README.md ├── .gitignore └── src/
Running Locally
Create and activate a virtual environment, then install the projectdependencies.
Run the MCP server:
fastmcp run server.py
The server uses the MCP stdio transport for local client connections.
Testing with MCP Inspector
The server can be inspected with FastMCP:
fastmcp inspect server.py
MCP Inspector can also be used to connect to the server and test theregistered tools.
Claude Desktop
The server can be connected to Claude Desktop as an MCP server.
Claude Desktop acts as the MCP client and can discover and invoke thearithmetic tools when required.
Example:
User: Calculate 25 × 8
Claude Desktop | v multiply(a=25, b=8) | v 200
Remote Deployment
The MCP server was deployed using Prefect Horizon.
Remote MCP endpoint:
https://claude-arithmetic-mcp.fastmcp.app/mcp
The deployed server provides the same four arithmetic tools through aremote MCP endpoint.
What I Learned
Basic MCP architecture and client-server communication
How to create and register MCP tools with FastMCP
How MCP clients discover and call tools
Testing MCP servers with MCP Inspector
Connecting an MCP server with Claude Desktop
Deploying a remote MCP server
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。