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

# Factory Droid

> Set up Factory AI's Droid coding agent

<Steps>
  <Step title="Install Droid CLI">
    <Tabs>
      <Tab title="macOS / Linux">
        ```bash theme={null}
        curl -fsSL https://app.factory.ai/cli | sh
        ```
      </Tab>

      <Tab title="Windows">
        ```powershell theme={null}
        irm https://app.factory.ai/cli/windows | iex
        ```
      </Tab>
    </Tabs>

    For the latest installation instructions and platform-specific guidance, visit the [Factory AI CLI documentation](https://factory.ai/product/cli).
  </Step>

  <Step title="Authenticate with Factory AI">
    To use Droid in Vibe Kanban, you must first authenticate:

    1. Run `droid` to launch the interactive CLI
    2. Use the `/login` command within Droid to authenticate

    Alternatively, generate an API key from [Factory Settings](https://app.factory.ai/settings/api-keys) and set it as an environment variable:

    ```bash theme={null}
    export FACTORY_API_KEY=fk-...
    ```

    For detailed authentication instructions, see the [Factory AI documentation](https://docs.factory.ai/factory-cli/getting-started/overview).
  </Step>

  <Step title="Start Vibe Kanban">
    Once authenticated, launch Vibe Kanban:

    ```bash theme={null}
    npx vibe-kanban
    ```

    You can now select Factory Droid when creating task attempts.
  </Step>
</Steps>

## Configuration Options

Droid supports several configuration options in Vibe Kanban:

* **Autonomy Level**: Controls permission level for file and system operations
  * `normal`: Read-only mode – safe for reviewing planned changes without execution
  * `low`: Basic file creation/editing while blocking system changes
  * `medium`: Development operations with recoverable side effects
  * `high`: Production operations with security implications or major side effects
  * `skip-permissions-unsafe`: Bypasses all permission checks (use only in isolated environments) (default)

* **Model**: Specify which model to use

* **Reasoning Effort**: Control the reasoning depth
  * `off`: Minimal reasoning
  * `low`: Light reasoning
  * `medium`: Balanced reasoning
  * `high`: Deep reasoning for complex tasks

These options can be configured when creating agent configurations in Vibe Kanban.