jupyter-mcp-server

Overview

The Jupyter MCP Server developed by Datalayer allows AI models to connect and manage Jupyter notebooks in real-time. It provides a comprehensive set of tools for notebook management, cell operations, and execution, with multimodal output support and JupyterLab integration. The server maintains context awareness across entire notebooks and works with any MCP client such as Claude Desktop, Cursor, and Windsurf.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• list_files

  List files and directories in the Jupyter server's file system.

  Try:Show me the files in the current directoryTry:List all notebooks in the project
• list_kernels

  List all available and running kernel sessions on the Jupyter server.

  Try:Show all running Jupyter kernelsTry:List available Python kernels
• connect_to_jupyter

  Connect to a Jupyter server dynamically without restarting the MCP server.

  Try:Connect to the Jupyter server at localhost:8888Try:Establish connection to production Jupyter
• use_notebook

  Connect to a notebook file, create a new one, or switch between notebooks.

  Try:Open my_analysis.ipynb notebookTry:Create a new notebook called results
• list_notebooks

  List all notebooks available on the Jupyter server and their status.

  Try:Show all available notebooksTry:List notebooks and their running status
• restart_notebook

  Restart the kernel for a specific managed notebook.

  Try:Restart the kernel for my_analysis.ipynbTry:Reset the notebook environment
• unuse_notebook

  Disconnect from a specific notebook and release its resources.

  Try:Close the current notebookTry:Disconnect from my_analysis.ipynb
• read_notebook

  Read notebook cells source content with brief or detailed format options.

  Try:Show me all cells in my_analysis.ipynbTry:Display notebook content with outputs
• read_cell

  Read the full content (Metadata, Source and Outputs) of a single cell.

  Try:Show me cell 5 contentsTry:Display the content of the selected cell
• insert_cell

  Insert a new code or markdown cell at a specified position.

  Try:Add a new code cell after cell 3Try:Insert a markdown cell at the beginning
• delete_cell

  Delete a cell at a specified index.

  Try:Remove cell 7Try:Delete the selected cell
• execute_cell

  Execute a cell with timeout, supports multimodal output including images.

  Try:Run cell 4Try:Execute the selected cell and show results

Comparable tools

Installation

# Install required dependencies
pip install jupyterlab==4.4.1 jupyter-collaboration==4.0.2 jupyter-mcp-tools>=0.1.4 ipykernel pycrdt

# Start JupyterLab
jupyter lab --port 8888 --IdentityProvider.token MY_TOKEN --ip 0.0.0.0

{
  "mcpServers": {
    "jupyter": {
      "command": "uvx",
      "args": ["jupyter-mcp-server@latest"],
      "env": {
        "JUPYTER_URL": "http://localhost:8888",
        "JUPYTER_TOKEN": "MY_TOKEN",
        "ALLOW_IMG_OUTPUT": "true"
      }
    }
  }
}

FAQ

What Jupyter deployments are compatible with this MCP server?

The server is compatible with any Jupyter deployment including local servers, JupyterHub, and cloud platforms. It also works with Datalayer hosted notebooks.

How does the server handle notebook context awareness?

The server maintains context awareness across entire notebooks by understanding the complete notebook structure, cell relationships, and execution history to provide more relevant AI interactions.

Compare jupyter-mcp-server with