Home page

Running marimo edit without a filename opens the home page, which lets you browse and manage notebooks in a directory.

Usage

bash

# Open home page for current directory
marimo edit

# Open home page for a specific folder
marimo edit folder/

The home page shows all marimo notebooks in the directory, letting you:

Open existing notebooks
Create new notebooks
See notebook metadata

Sandboxed Home

You can run the home page in sandbox mode (called "Sandboxed Home"), where each notebook gets its own isolated environment:

bash

marimo edit --sandbox folder/

When using Sandboxed Home:

Each notebook runs in its own isolated environment
Dependencies are read from each notebook's inline script metadata (PEP 723)
Environments are created on-demand when you open a notebook

This is useful when you have a collection of notebooks with different dependencies and want to keep them isolated from each other.

!!! note "Additional dependencies required"

Sandboxed Home requires additional packages:

```bash
uv add 'marimo[sandbox]'
```

This installs `pyzmq` (for inter-process communication) and `uv`
(for environment management).

Using custom virtual environments

When using Sandboxed Home, you can specify an existing virtual environment for a notebook instead of having marimo create one automatically. This is configured using [tool.marimo.venv] in your script metadata:

python

# /// script
# [tool.marimo.venv]
# path = "path/to/venv"      # relative or absolute path
# writable = false           # optional, default is false
# ///

!!! note "Sandboxed Home only"

The `[tool.marimo.venv]` configuration only applies when using
Sandboxed Home (`marimo edit --sandbox folder/`). For single notebooks,
activate your virtual environment before running marimo:

```bash
source path/to/venv/bin/activate
marimo edit notebook.py
```

Configuration options

Option	Description
`path`	Path to the virtual environment (relative or absolute)
`writable`	Whether marimo can install packages into the venv (default: `false`)

Behavior

`writable`	marimo installed?	What happens
`true`	-	marimo installs itself and required dependencies into the venv
`false`	Yes	Uses the venv as-is (warns if marimo version differs)
`false`	No	Injects `PYTHONPATH` for marimo (requires matching Python version)

This is useful when:

You have a conda or poetry environment you want to reuse
You're working in a team with a shared environment
You want notebooks in a folder to use different pre-configured environments