jupyter-notebook-mcpA Model Context Protocol (MCP) for Jupyter Notebook

— (0)

Reviewed by Daniel Nikulshyn·Updated July 2026

Overview

JupyterMCP connects Jupyter Notebook to Claude AI through the Model Context Protocol (MCP), enabling AI-assisted code execution, data analysis, visualization, and more. This integration allows Claude to directly interact with and control Jupyter Notebooks. It's compatible only with Jupyter Notebook version 6.x and consists of three main components: a WebSocket server, a client JavaScript, and an MCP server.

Key features

Two-way communication through a WebSocket-based server
Cell manipulation (insert, execute, manage)
Notebook management (save, retrieve information)
Cell execution (run specific cells or all cells)
Output retrieval with text limitation options

Pricing

Model: Free
Category: MCP Servers
Rating: No reviews yet

Use cases

AI-assisted Data Analysis

Use Claude AI to interact with Jupyter Notebook for data analysis, visualization, and code execution.

Automated Code Execution

Leverage Claude AI to execute specific cells or all cells in a Jupyter Notebook, streamlining workflows.

Pros & Cons

Pros

Enables AI-assisted code execution and data analysis
Allows Claude AI to directly interact with and control Jupyter Notebooks
Supports two-way communication between Claude AI and Jupyter Notebook

Cons

Only compatible with Jupyter Notebook version 6.x
Requires Python 3.12 or newer and uv package manager
Setup process involves multiple steps and configuration

Battle record

Across 1 battle in the Pantheon.

1st

2nd

3rd

Last battle

#3
MCP Servers Showdown — July 6, 2024
Jul 6, 2024 · #3 of 4

Reviews

No reviews yet. Be the first!

Q&A

What are the prerequisites for installing JupyterMCP?

You need Python 3.12 or newer and the uv package manager. After cloning the repo, create a virtual environment, install the kernel, and optionally add packages like seaborn. Finally, configure the Claude desktop app with the mcpServers settings.

Asked by Thandiwe Dlamini · Jan 12, 2026

How does Claude AI interact with the notebook using JupyterMCP?

Claude can insert, execute, and manage cells, save notebooks, retrieve information, run specific or all cells, and fetch output content (with text limits) via a WebSocket-based two‑way communication channel.

Asked by Farah Rahimi · Dec 15, 2025

What Jupyter versions does JupyterMCP support?

JupyterMCP works only with Jupyter Notebook 6.x. It is not compatible with Jupyter Lab, Notebook v7.x, VS Code Notebooks, or Google Colab.

Asked by Ximenez Alvarado · Nov 16, 2025

What are the main components of the JupyterMCP integration?

It consists of three parts: a WebSocket server (jupyter_ws_server.py) inside the notebook, a client JavaScript (client.js) that runs in the notebook, and an MCP server (jupyter_mcp_server.py) that implements the Model Context Protocol.

Asked by Nils Johansson · Nov 18, 2025