Jupyter MCP Server

@datalayer

Repository Homepage

Basic information

About Jupyter MCP Server

🪐 🔧 Model Context Protocol (MCP) Server for Jupyter.

#tools #ai #jupyter #mcp #mcp-server

Jupyter MCP Server config

No standard config provided

This server doesn't expose a parseable MCP config block in its README. See the repository for install instructions.

Repository

Jupyter MCP Server tools

Tool list not available yet

This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is 🪐✨ Jupyter MCP Server?

An open-source MCP server that enables AI agents to connect to and manage Jupyter Notebooks in real-time. It bridges AI assistants with Jupyter environments, allowing direct notebook creation, cell execution, and output retrieval. Developed by Datalayer.

How to use 🪐✨ Jupyter MCP Server?

Install via pip install jupyter-mcp-server and configure your MCP client with environment variables JUPYTER_URL, JUPYTER_TOKEN, and ALLOW_IMG_OUTPUT. Run using uvx for quick start or Docker for production. Alternatively, run as a Jupyter Server extension. The server supports STDIO and Streamable HTTP transports.

Key features of 🪐✨ Jupyter MCP Server

Real-time notebook control with instant output viewing.
Smart execution that adjusts after cell failures.
Context-aware understanding of entire notebooks.
Multimodal output support (images, plots, text).
Multi-notebook switching and management.
JupyterLab integration with additional commands.

Use cases of 🪐✨ Jupyter MCP Server

AI-assisted code development and debugging in Jupyter.
Automated data analysis and visualization workflows.
Interactive exploration of notebooks via natural language.
Orchestrating multi-notebook experiments with an AI agent.
Real-time collaborative editing with JupyterLab integration.

Frequently asked questions

What MCP clients are supported?

Works with any MCP client, including Claude Desktop, Cursor, Windsurf, and others.

How do I configure the server?

Set `JUPYTER_URL`, `JUPYTER_TOKEN`, and `ALLOW_IMG_OUTPUT` environment variables. A `MCP_TOKEN` is also required since version 1.0.0.

Is Docker supported?

Yes. Use the `datalayer/jupyter-mcp-server` Docker image with environment variables passed at runtime.

What are the runtime requirements?

Python environment with JupyterLab 4.4.1+, `jupyter-collaboration`, `jupyter-mcp-tools`, `ipykernel`, and `pycrdt`. Starting v1.0.2, `datalayer_pycrdt` is no longer needed.

Does it support JupyterHub or Google Colab?

Support for JupyterHub and Google Colab deployments is actively being developed; user feedback is welcome.

Comments

More Data & Analytics MCP servers

Ask Xverum

Xverum-LLC

Find and enrich the right people and companies in plain English. Built for AI agents: 750M professional profiles, 54M company profiles, 200M+ verified business emails and phone numbers, and job-change scores with Next Mo

Tyton MCP

usetyton

Meta Pixel not working, or you need a Meta tracking audit? Tyton MCP finds the break and sets up Meta Pixel plus the Conversions API. https://usetyton.com

AudD Music Recognition

AudD

AudD MCP is a way for agents to add music recognition to apps or to fulfill song identification requests. It connects you to AudD, the music recognition API.

Subtext

Subtext by Fullstory

Session replay, built for agents. Subtext is agentic session review: it captures production sessions of your app and connects them to your coding agent — Claude Code, Cursor, Codex, Devin, your own harness — so it can

looot

One key and one balance for 2,350 data APIs from 76 providers your AI agent can call: email finding, enrichment, SEO, scraping, social data. Pay per call, price shown before every run.