Jupyter Server
An MCP interface/extension for Jupyter Server
What is Jupyter Server?
Jupyter Server is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to mcp interface/extension for jupyter server
An MCP interface/extension for Jupyter Server
This server falls under the Developer Tools category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- An MCP interface/extension for Jupyter Server
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx jupyter-serverConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Jupyter Server
jupyter-server-mcp is an official MCP extension for Jupyter Server that bridges AI assistants and live Jupyter notebook environments, enabling tools like Claude to read and edit notebook cells, execute code, manage kernels, perform Git operations, and issue JupyterLab commands — all without leaving the AI chat interface. It installs as a standard Jupyter extension and dynamically registers any Python function as an MCP tool, so teams can customize which capabilities they expose. Data scientists and ML engineers use it to run experiments, inspect dataframes, and iterate on notebooks through natural language prompts.
Prerequisites
- Python 3.9 or later with pip or conda
- A running Jupyter Server or JupyterLab instance
- An MCP-compatible client such as Claude Desktop or Cursor
- uvx (from the uv package) for the recommended stdio proxy connection method
- No external API keys required
Install jupyter-server-mcp
Install the extension into your Python environment using pip. Conda and pixi installations are also supported.
pip install jupyter-server-mcpConfigure the Jupyter extension (optional)
Create a jupyter_config.py in your working directory to specify the MCP server name, port, and which tools to expose. If omitted, the extension uses sensible defaults.
# jupyter_config.py
c = get_config()
c.MCPExtensionApp.mcp_name = "My Jupyter MCP Server"
c.MCPExtensionApp.mcp_port = 3001
c.MCPExtensionApp.mcp_tools = [
"os:getcwd",
"jupyter_ai_tools.toolkits.notebook:read_notebook",
"jupyterlab_commands_toolkit.tools:list_all_commands",
]Start Jupyter Lab with the MCP extension
Launch JupyterLab. The MCP extension will start automatically and listen on the configured port.
jupyter lab --config=jupyter_config.pyConfigure your MCP client (stdio proxy — recommended)
Add the stdio proxy to your claude_desktop_config.json. The proxy auto-discovers the running Jupyter Server — no URL configuration needed.
{
"mcpServers": {
"jupyter-mcp": {
"command": "uvx",
"args": ["--from", "jupyter-server-mcp", "jupyter-server-mcp-proxy"]
}
}
}Or connect via direct HTTP transport
If you set a fixed port in your Jupyter config, you can connect directly over HTTP instead of using the stdio proxy.
{
"mcpServers": {
"jupyter-mcp": {
"type": "http",
"url": "http://localhost:3001/mcp"
}
}
}Jupyter Server Examples
Client configuration (Claude Desktop — stdio proxy)
Recommended claude_desktop_config.json entry using the stdio proxy which auto-discovers the running Jupyter Server.
{
"mcpServers": {
"jupyter-mcp": {
"command": "uvx",
"args": ["--from", "jupyter-server-mcp", "jupyter-server-mcp-proxy"]
}
}
}Prompts to try
Example prompts that use the Jupyter MCP tools for notebook operations and data science workflows.
- "List all open notebooks in my Jupyter session."
- "Read the contents of the notebook analysis.ipynb and summarize what each cell does."
- "Add a new cell to analysis.ipynb that plots the DataFrame df as a bar chart."
- "Run git status in the current Jupyter working directory."
- "Clear all outputs from the notebook model_training.ipynb."Troubleshooting Jupyter Server
The stdio proxy cannot find a running Jupyter Server
Make sure Jupyter Lab is already running before starting the MCP client. The proxy discovers the server via Jupyter's runtime directory. If using a custom port, set the JUPYTER_SERVER_MCP_URL environment variable to point to the correct endpoint.
Tools are missing or the list is empty after connecting
The extension only exposes tools that are explicitly listed in MCPExtensionApp.mcp_tools in your jupyter_config.py. Add the specific tool module paths you need, then restart Jupyter Lab.
Permission errors when running Git operations through the MCP tools
The Jupyter Server process must have read/write access to the repository. Ensure you launched jupyter lab from within the Git repo directory, and that your Git identity (user.name, user.email) is configured globally.
Frequently Asked Questions about Jupyter Server
What is Jupyter Server?
Jupyter Server is a Model Context Protocol (MCP) server that mcp interface/extension for jupyter server It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Jupyter Server?
Follow the installation instructions on the Jupyter Server GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Jupyter Server?
Jupyter Server works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Jupyter Server free to use?
Yes, Jupyter Server is open source and available under the BSD-3-Clause license. You can use it freely in both personal and commercial projects.
Jupyter Server Alternatives — Similar Developer Tools Servers
Looking for alternatives to Jupyter Server? Here are other popular developer tools servers you can use with Claude, Cursor, and VS Code.
Ecc
★ 188.2kThe agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Javaguide
★ 155.8kJava 面试 & 后端通用面试指南，覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发
Gemini CLI
★ 104.5kA secure MCP server that wraps the Google Gemini CLI, allowing clients to query Gemini models using local OAuth sessions without requiring an API key. It provides tools for model interaction and diagnostics with built-in protection against command in
Awesome MCP Servers
★ 87.3k⭐ Curated list of Model Context Protocol (MCP) servers - tools that extend Claude Desktop, Cursor, Windsurf, and other MCP clients with custom capabilities.
MCP Servers
★ 86.0kModel Context Protocol Servers
CC Switch
★ 77.5kA cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Gemini CLI & Hermes Agent. Only official website: ccswitch.io
Browse More Developer Tools MCP Servers
Explore all developer tools servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Jupyter Server in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Jupyter Server?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.