Calculator MCP Server
A simple MCP server that provides basic calculator functionality for performing mathematical operations. Built with FastMCP and demonstrates fundamental MCP server implementation patterns.
README
Test Remort Server
This small project demonstrates a simple MCP server built with FastMCP.
You are using the uv package manager, so the instructions below show how to install dependencies and run the server with uv.
At a glance
- Main code file:
main.py - Lock file (pinned dependencies):
uv.lock - Package manager:
uv(Astral's uv)
Requirements
- Python >= 3.12 (see
pyproject.toml) uvinstalled on your system —uvis a lightweight modern Python environment & package manager. Installation: https://docs.astral.sh/uv/getting-started/installation/
Note: If uv is not installed, follow the link above. uv typically manages both the virtual environment and package installation for the project.
Install dependencies (using uv)
Run this from the project root (where uv.lock and pyproject.toml live):
uv sync
uv sync will read uv.lock and install the pinned dependency versions into the environment uv manages.
If you prefer not to use uv, you can create a normal Python virtual environment and use pip to install packages manually — however, using uv.lock via uv is recommended for reproducible installs.
Run the server
From the project root you can run the server inside uv's environment like this:
uv run python main.py
This runs the project's main.py using the Python environment managed by uv.
By default main.py binds the server to:
- Host:
127.0.0.1 - Port:
5000
Open your browser or curl to:
http://127.0.0.1:5000
Check the terminal output from uv run python main.py for available routes or log messages.
Useful uv commands (reference)
uv sync— Install/sync dependencies fromuv.lockuv run <command>— Run a command inside uv's managed environment (e.g.uv run pytest)uv pip install <pkg>— Run pip inside uv's environment to install a package (e.g.uv pip install fastmcp)
Troubleshooting
- If
uvis not found (command not found), make sure you've installed it and that it's on your PATH. Installation guide: https://docs.astral.sh/uv/getting-started/installation/ - If
uv syncfails, check the terminal output; common causes are network / SSL issues or missing system build tools (e.g.build-essential,python3-dev,libssl-devon Linux).
Next steps (optional)
- If you'd like, I can expose a small ASGI
appinmain.pyso you can also run the project via uvicorn, for example:
uv run uvicorn main:app --host 127.0.0.1 --port 5000
Tell me if you want that change and I will update main.py and add a couple of example curl requests.
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。