Jupyter MCP Server
Jupyter MCP Server is a Model Context Protocol (MCP) server implementation that provides interaction with Jupyter notebooks running in any JupyterLab (works also with your local JupyterLab).
Jupyter MCP Server
Start JupyterLab
Make sure you have the following installed. The collaboration package is needed as the modifications made on the notebook can be seen thanks to Jupyter Real Time Collaboration.
Then, start JupyterLab with the following command.
You can also run
make jupyterlab.[!NOTE]The--ipis set to0.0.0.0to allow the MCP server running in a Docker container to access your local JupyterLab.
Use with Claude Desktop
Claude Desktop can be downloaded from this page for macOS and Windows.
For Linux, we had success using this UNOFFICIAL build script based on nix
To use this with Claude Desktop, add the following to your
claude_desktop_config.json (read more on the MCP documentation website).[!IMPORTANT]Ensure the port of theSERVER_URLandTOKENmatch those used in thejupyter labcommand.TheNOTEBOOK_PATHshould be relative to the directory where JupyterLab was started.
Claude Configuration on macOS and Windows
Claude Configuration on Linux
Components
Tools
The server currently offers 2 tools:
add_execute_code_cell
- Add and execute a code cell in a Jupyter notebook.
- Input:
- Returns: Cell output.
add_markdown_cell
- Add a markdown cell in a Jupyter notebook.
- Input:
- Returns: Success message.
Building
You can build the Docker image it from source.
Installing via Smithery
To install Jupyter MCP Server for Claude Desktop automatically via Smithery:
Jupyter MCP Server
Jupyter MCP Server is a Model Context Protocol (MCP) server implementation that provides interaction with Jupyter notebooks running in any JupyterLab (works also with your local JupyterLab).
Jupyter MCP Server
Start JupyterLab
Make sure you have the following installed. The collaboration package is needed as the modifications made on the notebook can be seen thanks to Jupyter Real Time Collaboration.
Then, start JupyterLab with the following command.
You can also run
make jupyterlab.[!NOTE]The--ipis set to0.0.0.0to allow the MCP server running in a Docker container to access your local JupyterLab.
Use with Claude Desktop
Claude Desktop can be downloaded from this page for macOS and Windows.
For Linux, we had success using this UNOFFICIAL build script based on nix
To use this with Claude Desktop, add the following to your
claude_desktop_config.json (read more on the MCP documentation website).[!IMPORTANT]Ensure the port of theSERVER_URLandTOKENmatch those used in thejupyter labcommand.TheNOTEBOOK_PATHshould be relative to the directory where JupyterLab was started.
Claude Configuration on macOS and Windows
Claude Configuration on Linux
Components
Tools
The server currently offers 2 tools:
add_execute_code_cell
- Add and execute a code cell in a Jupyter notebook.
- Input:
- Returns: Cell output.
add_markdown_cell
- Add a markdown cell in a Jupyter notebook.
- Input:
- Returns: Success message.
Building
You can build the Docker image it from source.
Installing via Smithery
To install Jupyter MCP Server for Claude Desktop automatically via Smithery: