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

# Scripts

> How Superscale writes your ad scripts and storyboards, and how to steer them.

Every effective video starts with a script. Superscale turns a brief and research context into a scene-by-scene storyboard before video generation, so the hook, message, visuals, and voiceover are decided before credits are spent.

## Where scripts come from

<CardGroup cols={2}>
  <Card title="Built from research">
    The hook and structure can draw on what is working in your niche: competitor research, saved references, or account learnings.
  </Card>

  <Card title="Grounded in your context">
    Claims, tone, and product details come from your [brand and products](/context/overview).
  </Card>

  <Card title="Scene by scene">
    The script is a storyboard, with each scene's visual and voiceover mapped out before anything renders.
  </Card>

  <Card title="Yours to steer">
    Guide the angle, tighten a line, or ask for a different hook, then regenerate.
  </Card>
</CardGroup>

## Guide a script

Name the audience, angle, and intended outcome. Example: "lead with the time-saving benefit, end on the free trial."

## Script versus post-processing

Keep the storyboard focused on what the camera sees and what the viewer hears. Some elements belong outside the generated scene:

| Element                                      | Where it belongs                                                                     |
| -------------------------------------------- | ------------------------------------------------------------------------------------ |
| Spoken lines in English native-audio formats | Inside the scene descriptions and mirrored as the transcript.                        |
| Non-English voiceover                        | In the voiceover field, mixed in after render.                                       |
| Captions and subtitles                       | Post-processing caption settings, not Seedance scene text.                           |
| End cards                                    | End-card prompt or exact end-card source file, appended after the video.             |
| Logo holds, lower thirds, dense overlays     | Edit/export layer where possible; do not rely on the video model to draw exact text. |

<Note>
  For voiceover-mix formats, the scene descriptions should stay visual and ambient. Put the spoken script in the voiceover, not in each scene.
</Note>

<Card title="Get better hooks" href="/best-practices/creation" horizontal>
  Tips for scripts, hooks, and angles.
</Card>