jupyterlab_commands_toolkit

A Jupyter extension that provides an AI toolkit for JupyterLab commands.

This extension is composed of a Python package named jupyterlab_commands_toolkit for the server extension and a NPM package named jupyterlab-commands-toolkit for the frontend extension.

Features

• Command Discovery: List all available JupyterLab commands with their metadata
• Command Execution: Execute any JupyterLab command programmatically from Python
• MCP Integration: Automatically exposes tools to AI assistants via jupyter-server-mcp
• Web Client Routing: Address a command to a specific browser tab so it runs only there (see below)

Web client routing

By default a lab_command is broadcast to every connected browser, so an AI-driven command (for example, running a notebook cell) runs on all of them. To target one client, the event carries an optional client_id: the frontend runs the command only when it matches this tab's web client id, and ignores it otherwise. A command with no client_id still runs everywhere, so existing behavior is unchanged when nothing opts in.

• Each browser tab has a stable web_client_id, provided via the IWebClientId token and queryable through the jupyterlab-commands-toolkit:get-web-client-id command.
• When @jupyter/chat is installed, the toolkit stamps this id into the metadata of every message the tab sends, so a server-side agent can learn which client triggered a message. This integration is optional and is a no-op when Jupyter Chat is absent.
• On the server, set the jupyterlab_commands_toolkit.tools.target_client_id contextvar (an MCP middleware does this from a request header) and execute_command stamps it onto the emitted event as client_id.

See ui-tests/README.md for the end-to-end suites covering each of these paths.

Requirements

• JupyterLab >= 4.5.0a3

Install

To install the extension, execute:

pip install jupyterlab_commands_toolkit

To install with jupyter-server-mcp integration support:

pip install jupyterlab_commands_toolkit[mcp]

Usage

With jupyter-server-mcp (Recommended)

This extension automatically registers its tools with jupyter-server-mcp via Python entrypoints, making them available to AI assistants and other MCP clients.

1. Install both packages:

pip install jupyterlab_commands_toolkit[mcp]

2. Start Jupyter Lab (the MCP server starts automatically):

jupyter lab

3. Configure your MCP client (e.g., Claude Desktop) to connect to http://localhost:3001/mcp

The following tools will be automatically available:

• list_all_commands - List all available JupyterLab commands with their metadata
• execute_command - Execute any JupyterLab command programmatically

Server-Side Python Usage

Use the toolkit directly from server-side Python to execute JupyterLab commands. These functions must run in the initialized Jupyter Server process, such as via jupyter-server-mcp or another server extension.

from jupyterlab_commands_toolkit.tools import execute_command, list_all_commands

async def main():
    # List all available commands
    commands = await list_all_commands()

    # Toggle the file browser
    result = await execute_command("filebrowser:toggle-main")

    # Run notebook cells
    result = await execute_command("notebook:run-all-cells")

For a full list of available commands in JupyterLab, refer to the JupyterLab Command Registry documentation.

Uninstall

To remove the extension, execute:

pip uninstall jupyterlab_commands_toolkit

Troubleshoot

If you are seeing the frontend extension, but it is not working, check that the server extension is enabled:

jupyter server extension list

If the server extension is installed and enabled, but you are not seeing the frontend extension, check the frontend extension is installed:

jupyter labextension list

Contributing

Development install

Note: You will need NodeJS to build the extension package.

The jlpm command is JupyterLab's pinned version of yarn that is installed with JupyterLab. You may use yarn or npm in lieu of jlpm below.

# Clone the repo to your local environment
# Change directory to the jupyterlab_commands_toolkit directory
# Install package in development mode
pip install -e "."
# Link your development version of the extension with JupyterLab
jupyter labextension develop . --overwrite
# Server extension must be manually installed in develop mode
jupyter server extension enable jupyterlab_commands_toolkit
# Rebuild extension Typescript source after making changes
jlpm build

You can watch the source directory and run JupyterLab at the same time in different terminals to watch for changes in the extension's source and automatically rebuild the extension.

# Watch the source directory in one terminal, automatically rebuilding when needed
jlpm watch
# Run JupyterLab in another terminal
jupyter lab

With the watch command running, every saved change will immediately be built locally and available in your running JupyterLab. Refresh JupyterLab to load the change in your browser (you may need to wait several seconds for the extension to be rebuilt).

By default, the jlpm build command generates the source maps for this extension to make it easier to debug using the browser dev tools. To also generate source maps for the JupyterLab core extensions, you can run the following command:

jupyter lab build --minimize=False

Development uninstall

# Server extension must be manually disabled in develop mode
jupyter server extension disable jupyterlab_commands_toolkit
pip uninstall jupyterlab_commands_toolkit

In development mode, you will also need to remove the symlink created by jupyter labextension develop command. To find its location, you can run jupyter labextension list to figure out where the labextensions folder is located. Then you can remove the symlink named jupyterlab-commands-toolkit within that folder.

Packaging the extension

See RELEASE