Livedocs Anywhere lets you run notebooks on your own machine while using the familiar Livedocs web editor. Your code executes locally with full access to local files, databases, and compute resources.
Why Run Locally?
- Access local data — Connect to databases on your machine or network, read files from your filesystem
- Use your compute — Run intensive workloads on your hardware instead of cloud VMs
- Work offline — Continue working without an internet connection (after initial setup)
- Security requirements — Keep sensitive data on-premises when compliance requires it
Installation
Requirements
Before installing, you need uv (a fast Python package manager). If you don’t have it:
curl -LsSf https://astral.sh/uv/install.sh | sh
Install the CLI
Run the installer:
curl -fsSL https://livedocs.com/install.sh | bash
The installer will:
- Download the latest CLI bundle for your platform (macOS or Linux)
- Extract it to
~/.livedocs - Prompt to add the CLI to your PATH
To skip the prompt and auto-configure your PATH:
curl -fsSL https://livedocs.com/install.sh | bash -s -- --setup-path
Verify Installation
Open a new terminal (or run source ~/.zshrc) and check:
livedocs --version
Quick Start
1. Pair with your workspace
livedocs setup
This opens your browser to authenticate with Livedocs. Once you approve, your workspace token is stored securely in your system keychain.
If you’re on a headless server or prefer to use an API token:
livedocs setup --api-token YOUR_TOKEN
You can generate API tokens from your workspace settings at livedocs.com.
2. Launch a notebook
livedocs launch
You’ll see a list of your recent notebooks. Select one, or choose to create a new notebook.
To launch a specific notebook directly:
livedocs launch --notebook NOTEBOOK_ID
To create and launch a new notebook:
livedocs launch --new
The CLI starts the runtime and opens the editor in your browser. You’re now running locally.
3. Stop the session
Press Ctrl+C in the terminal, or run:
livedocs stop
How It Works
When you run livedocs launch:
- The CLI authenticates with Livedocs using your workspace token
- Livedocs allocates a session and returns connection credentials
- The CLI bootstraps a Python environment locally using uv
- Two processes start on your machine:
- Middleman — the notebook runtime that executes your code
- Jedi — the AI assistant service
- Your browser opens the Livedocs editor, connected to your local runtime
The web interface handles editing, collaboration, and notebook storage. The local runtime handles code execution with full access to your machine.
What stays local
- Code execution happens entirely on your machine
- File I/O reads and writes to your local filesystem
- Database connections go directly from your machine
- The AI assistant runs locally (queries are sent to AI providers)
What goes through Livedocs
- Notebook content (cells, outputs) syncs with Livedocs for storage
- Authentication and session management
- Collaboration features (if enabled)
Importing Jupyter Notebooks
You can import existing .ipynb files when creating a new notebook:
livedocs launch --new --import path/to/notebook.ipynb
This creates a new Livedocs notebook and imports the cells from your Jupyter file.
Running Detached
By default, sessions run in the background. You can close the terminal and the runtime keeps running.
To see logs while launching:
livedocs launch --attach
To check on a running session:
livedocs status
To follow logs from a running session:
livedocs logs
Next Steps
- See the full Commands Reference for all available options
- Check Troubleshooting if you run into issues