jjsantos01/jupyter-notebook-mcp

Overview

A Model Context Protocol (MCP) server that connects AI assistants to Jupyter Notebook environments. It allows agents to execute code, create and modify notebooks, and inspect outputs through a standardized interface.

Best for

Best for
Developers building AI agents that need to execute and inspect Python code in Jupyter notebooks

Use cases

• Automating notebook creation and cell execution from AI prompts
• Running data analysis scripts in Jupyter via an MCP client
• Integrating Jupyter workflows into agent-based development tools

How to use

Install

uv pip install seaborn

Tools exposed

• uv
• git
• python
• pip
• ipykernel
• stata-setup

Tested with

Claude Desktop, VS Code

Example client config

{\n    "mcpServers": {\n        "jupyter": {\n            "command": "uv",\n            "args": [\n                "--directory",\n                "/ABSOLUTE/PATH/TO/PARENT/REPO/FOLDER/src",\n                "run",\n                "jupyter_mcp_server.py"\n            ]\n        }\n    }\n}

Notes

A Model Context Protocol (MCP) server that connects AI assistants to Jupyter Notebook environments. It allows agents to execute code, create and modify notebooks, and inspect outputs through a standardized interface.

129 stars on GitHub. Last updated 2025-04-02. Licensed MIT.

Use cases

• Automating notebook creation and cell execution from AI prompts
• Running data analysis scripts in Jupyter via an MCP client
• Integrating Jupyter workflows into agent-based development tools

Pros

• Enables direct AI interaction with live Jupyter kernels
• Lightweight and focused on a single, well-defined protocol
• Open source with a growing community (129 stars)

Cons

• Requires an MCP-compatible client to be useful
• Limited to Jupyter Notebook environments only
• Relatively new project with fewer integrations and documentation

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.