> ## Documentation Index
> Fetch the complete documentation index at: https://mulerun.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Quickstart

> Install the MuleRun CLI, log in, and run your first coding-agent and Studio commands

This guide walks through the full happy path: install, log in, launch the
coding agent, generate an image with Studio.

<Steps>
  <Step title="Install">
    ```bash theme={null}
    npm install -g @mulerunai/cli
    mulerun --version
    ```

    See [Installation](/cli/installation) for `bun`, `pnpm`, and one-off
    `npx` usage.
  </Step>

  <Step title="Log in">
    ```bash theme={null}
    mulerun login
    ```

    A browser window opens and walks you through OAuth. The resulting token
    is cached locally; every other command reuses it.
  </Step>

  <Step title="Generate an image with Studio">
    ```bash theme={null}
    mulerun studio run openai/gpt-image-2/generation \
      --prompt "A vector logo of a fox, flat style" \
      --size "1024x1024"
    ```

    `mulerun studio` is the multimodal generation surface — images, videos,
    speech, music. See [`mulerun studio`](/cli/commands/studio) for the full
    catalog.
  </Step>
</Steps>

## What you got

Most commands print pretty text by default and JSON when you add `-o json`.
Streaming responses (from `code`, `session`, `computer chat`) go to stdout
as they arrive; status messages go to stderr.

For coding work, `mulerun code` launches an interactive coding-agent CLI
under the same login — see [`mulerun code`](/cli/commands/code).

## Next

* Read the [command reference](/cli/commands/code) — start with `code` and
  `studio`, then explore session, computer, drive, page.
* Skim [Configuration](/cli/configuration) for `.env`, proxies, and
  per-command overrides.