MCP Server Demo
A demonstration project showing how to set up and use a custom MCP server with Claude Desktop, providing example tools and resources.
README
MCP Server Setup Guide
This project demonstrates how to set up and use a custom MCP server with Claude Desktop. The server provides example tools and resources that can be called from Claude Desktop or the MCP Inspector.
1. Install UV (Python Package Manager)
UV is a fast, modern package manager for Python. Install it globally (not inside a virtual environment):
pip install uv
2. Create a New UV-Managed MCP Project
uv init mcp-server-demo
cd mcp-server-demo
3. Add MCP as a Project Dependency
uv add "mcp[cli]"
If you see the following image, your MCP development tools are installed as expected:

Screenshot: Successful MCP tools installation in the terminal.
4. Install and Run the Server
You can install this server in Claude Desktop and interact with it right away:
mcp install server.py
Or test it with the MCP Inspector:
mcp dev server.py
5. Using the Installed Server with Claude Desktop
Verifying Installation in Claude Desktop
-
Check the terminal for a success message after running
mcp install server.py:
Screenshot: Successful installation message in the terminal. -
Check the MCP config file in Claude Desktop to verify that the JSON has been updated:
Screenshot: MCP config file updated with your server. -
Open Claude Desktop (restart if it was already running, for the updated config to take effect).
-
In the chat, try using your tool (e.g.,
add 2 3) and verify it is using theaddtool from your MCP server:
Screenshot: Using the custom tool from Claude Desktop chat.
Troubleshooting
- If your server does not appear in Claude Desktop, ensure you restarted the app after installation.
- Double-check that
mcp install server.pycompleted successfully and the config file was updated. - If you update your server code, reinstall it with
mcp install server.pyand restart Claude Desktop.
To extend this demo, edit the Python files as needed! For questions, reach out to kannan.ramaswamy01@sap.com
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。