---
title: Write video prompts
display: 'Write video *prompts*'
nav_title: Prompt video
summary:
  - "Let the first frame carry the look, and spend the video prompt on what changes during the shot."
  - "Write the shot in six parts: start state, one action, end state, camera, pace and sound."
  - "For a longer clip, give each beat its time, or write each shot on its own line and start the next one with Cut to."
description: Write video prompts in six parts (start state, one action, end state, camera, pace, sound), then time the beats of a longer clip or cut it into shots.
section: methods
order: 30
type: guide
level: Build
roles: [creator, artist, art-director, marketer, team-lead, developer]
time: { read: 10, do: 20, wait: 5 }
last_verified: 2026-11-11
models_checked: ["Seedance 2.5", "Seedance 2.5 Draft Complete", "Minimax H3", "Gemini Omni 1.1 Flash", "Wan 3.0", "Kling V3 I2V Pro"]
hero: methods/prompt-video/kitchen-first-frame.webp
related: [generate-video, direct-the-camera-and-motion, reference-to-video]
video: tutorials/prompt-video.mp4
video_caption: Write video prompts
about: "The film animates one first frame, a girl and her chameleon in a sunny kitchen, with Seedance 2.5. It writes the prompt in six parts, then turns it into a 15-second shot with four timed beats, then into three shots with cuts, and ends with the prompts that go wrong. The models and settings in this lesson are as of November 2026."
chapters:
  - { at: 0, title: "First frame", image: "/media/images/methods/prompt-video/ch-01-first-frame.webp" }
  - { at: 25.0, title: "Six parts", image: "/media/images/methods/prompt-video/ch-02-six-parts.webp" }
  - { at: 73.72, title: "Timed beats", image: "/media/images/methods/prompt-video/ch-03-timed-beats.webp" }
  - { at: 113.65, title: "Cuts", image: "/media/images/methods/prompt-video/ch-04-cuts.webp" }
  - { at: 146.36, title: "Tips", image: "/media/images/methods/prompt-video/ch-05-tips.webp" }
thumb: methods/prompt-video/thumb.webp
steps:
  - title: "Start from the first frame"
    text: "Create › Videos › Seedance 2.5, with the approved image in First Frame. The frame carries the girl, the chameleon, the kitchen and the light, so the prompt describes only what changes."
  - title: "Write the shot in six parts"
    text: "In order: start state, one main action, end state, camera (shot size and one exact move), pace and sound."
    prompt: |-
      The girl with the long white bubble braid stands still beside the kitchen table, looking at her green chameleon on the edge of the fruit bowl. She steps up to the table, picks up the lemon with both hands and holds it out to the chameleon, which slowly turns its head and both eyes toward the lemon. She ends standing at the table, holding the lemon up in front of the chameleon, smiling. Medium-long shot at table height, slow push-in toward the girl and the chameleon. Gentle, real-time pace. Sound: soft kitchen room tone, birds outside the window, light footsteps on the tiles. No music, no dialogue.
  - title: "Set the duration and generate"
    text: "Settings: Duration 8s, Resolution 480p, Aspect Ratio Auto, Generate Audio on, Output Format MP4, Draft on. Check the price on Generate and click it; Finalize in 1080p opens Seedance 2.5 Draft Complete with the draft loaded."
  - title: "Time the beats of a longer shot"
    text: "One opening line, then one action per beat on its own line with its time in brackets, one camera move for the whole shot, pace and sound. Duration 15s, the end of the last beat. The film shows the prompt as far as its sound line."
    prompt: |-
      One continuous shot, no cuts. The girl with the long white bubble braid stands beside the kitchen table, looking at her green chameleon on the edge of the fruit bowl.
      [0-4s] She steps up to the table, picks up the lemon with both hands and holds it out to the chameleon.
      [4-8s] The chameleon shoots out its long sticky tongue, grabs a green grape from the bowl and pulls it into its mouth. She laughs, surprised.
      [8-12s] She sets the lemon down right next to the chameleon. The chameleon's skin slowly changes color from green to the same bright lemon yellow, starting at its head.
      [12-15s] She leans in with her chin on her hands, smiling at the yellow chameleon, and it blinks at her.
      Camera: medium-long shot at table height. The camera slowly arcs to the right around the table and moves a little closer during the whole shot, ending on a medium shot of the girl and the chameleon.
      Pace: gentle, real-time.
      Sound: soft kitchen room tone, birds outside the window, a tiny wet snap when the tongue catches the grape.
  - title: "Ask for cuts"
    text: "Say what stays the same in every shot, then one line per shot with its time, framing and camera, each new one starting \"Cut to\"; Duration 12s, the end of the last shot. The film shows only the opening below; shot 3, a close-up with a slow push-in on her smile (8 to 12 s), follows the same pattern."
    prompt: |-
      Three shots with hard cuts. The same girl, the same chameleon, the same kitchen and the same sunlight in every shot.
      Shot 1 [0-4s]: medium-long shot at table height, static camera. The girl with the long white bubble braid stands beside the table, then steps up and picks up the lemon from the fruit bowl.
      Cut to shot 2 [4-8s]: extreme close-up of the green chameleon on the edge
agent_steps:
  - title: "Connect the Scenario MCP"
    text: "Connect your agent and sign in once at [mcp.scenario.com](https://mcp.scenario.com). Then tell it which Scenario team and project to use, so every file lands there."
  - title: "Give the agent the first frame"
    text: "The approved image of the girl and her chameleon in the sunny kitchen. Upload it with upload_asset or give its asset ID."
  - title: "Brief the agent: one shot"
    prompt: |-
      Animate the first frame with Seedance 2.5 (model_bytedance-seedance-2-5): the image as its first-frame image, no reference images, duration 8, aspect ratio following the frame, audio on, MP4, and its draft mode on for a 480p preview (read model_schema_get for the parameter names). Quote the cost with a dry run first and wait for my go.

      Prompt, word for word:
      The girl with the long white bubble braid stands still beside the kitchen table, looking at her green chameleon on the edge of the fruit bowl. She steps up to the table, picks up the lemon with both hands and holds it out to the chameleon, which slowly turns its head and both eyes toward the lemon. She ends standing at the table, holding the lemon up in front of the chameleon, smiling. Medium-long shot at table height, slow push-in toward the girl and the chameleon. Gentle, real-time pace. Sound: soft kitchen room tone, birds outside the window, light footsteps on the tiles. No music, no dialogue.
  - title: "Brief the agent: timed beats"
    prompt: |-
      Same first frame and settings, duration 15, the end of the last beat. Keep the prompt's line breaks. Quote it first and wait for my go.

      Prompt, word for word:
      One continuous shot, no cuts. The girl with the long white bubble braid stands beside the kitchen table, looking at her green chameleon on the edge of the fruit bowl.
      [0-4s] She steps up to the table, picks up the lemon with both hands and holds it out to the chameleon.
      [4-8s] The chameleon shoots out its long sticky tongue, grabs a green grape from the bowl and pulls it into its mouth. She laughs, surprised.
      [8-12s] She sets the lemon down right next to the chameleon. The chameleon's skin slowly changes color from green to the same bright lemon yellow, starting at its head.
      [12-15s] She leans in with her chin on her hands, smiling at the yellow chameleon, and it blinks at her.
      Camera: medium-long shot at table height. The camera slowly arcs to the right around the table and moves a little closer during the whole shot, ending on a medium shot of the girl and the chameleon.
      Pace: gentle, real-time.
      Sound: soft kitchen room tone, birds outside the window, a tiny wet snap when the tongue catches the grape.
  - title: "Brief the agent: three shots with cuts"
    text: "The article shows only the opening of this prompt, so the agent completes it from the outline, in the same pattern."
    prompt: |-
      Same first frame and settings, duration 12. Start the prompt with these lines, word for word, then finish shot 2 (the green chameleon turns from green to lemon yellow, 4 to 8 s) and add shot 3 on its own line, starting "Cut to shot 3 [8-12s]:", a close-up with a slow push-in on her smile. Show me the full prompt, quote it and wait for my go.

      Three shots with hard cuts. The same girl, the same chameleon, the same kitchen and the same sunlight in every shot.
      Shot 1 [0-4s]: medium-long shot at table height, static camera. The girl with the long white bubble braid stands beside the table, then steps up and picks up the lemon from the fruit bowl.
      Cut to shot 2 [4-8s]: extreme close-up of the green chameleon on the edge
  - title: "Check, finalize and file"
    text: "Accept when the clip starts on the first frame, every action happens once and in order, it ends on the end state, the camera makes the one move asked for, and the girl, chameleon and kitchen stay the same across cuts. Reject a repeated or new action, a clip that ends mid-action, a skipped beat, a different camera move, or a cut that brings in a new character or room."
    prompt: |-
      Wait for each job with jobs_wait and show me the drafts. Check each against the prompt beat by beat and tell me what fails; fix the prompt or the duration and quote a new draft for only those. Finalize the approved drafts in 1080p with Seedance 2.5 Draft Complete (find its model id with models_list), quoting first. Then create a collection with the first frame and the takes, and report the models, the settings and every asset ID.
updated: 2026-11-11
canonical: https://learn.scenario.com/methods/prompt-video
---

Last updated: 2026-11-11

# Write video prompts

> Write video prompts in six parts (start state, one action, end state, camera, pace, sound), then time the beats of a longer clip or cut it into shots.

In short:

- Let the first frame carry the look, and spend the video prompt on what changes during the shot.
- Write the shot in six parts: start state, one action, end state, camera, pace and sound.
- For a longer clip, give each beat its time, or write each shot on its own line and start the next one with Cut to.


A video prompt is not an image prompt that moves. The first frame already sets the look, so the prompt's job is to say what changes, how the camera follows it and what it sounds like. This lesson animates one approved image, a girl and her chameleon in a sunny kitchen, with Seedance 2.5: first as one shot written in six parts, then as a 15-second shot with four timed beats, then as three shots joined by cuts.

:::note As of November 2026
The models, settings and screens in this article are the ones Scenario offered in November 2026. New models and updates arrive often, so check the app for newer options before you choose.
:::

:::steps Quick version
1. Load an approved image as the **First Frame**, and let it carry the look.
2. Write the shot in six parts: start state, one main action, end state, camera, pace and sound.
3. In **Settings**, set a **Duration** that fits the action, then click **Generate**.
4. For a longer shot, give each beat its time, one action per beat, with one camera move for the whole shot.
5. For several shots in one clip, write each on its own line and start the next one with "Cut to".
:::

## Before you start

- **An approved first frame.** Make and approve the image before you animate it. The video keeps its characters, setting, light and style, so fix any problem in the image, not in the video prompt.
- **A model that takes a first frame.** The film uses **Seedance 2.5** in **Create › Videos**. Seedance 2.5 picks its mode from what you attach: a prompt alone, a first frame, a first and last frame, or references. A first frame and reference images can't be used together, so choose one anchor per shot.
- **Draft runs.** With **Draft** on in **Settings**, Seedance 2.5 renders a 480p preview. When a preview is the one you want, you finalize it in 1080p without generating it again.
- **What a video prompt holds.** Seedance's own format has slots for subject and action, scene, visual style, camera movement and audio. With a first frame, the scene and style are already in the image, which is why the prompt here spends its words on the action, the camera and the sound.

## Step 1: Start from the first frame

Start from an image you've already approved, and use it as the **First Frame**. It already shows the girl, her chameleon, the kitchen and the light. So the prompt doesn't describe the look again: it describes what changes.

![The video panel with Seedance 2.5, the prompt, and the kitchen image loaded as First Frame](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/ac9374b66d6f6eb1f09c625387cefa2478fe7912672a9453e7b9cc5f1ed0e4b2.webp)

Describing the look a second time does not help. At best it repeats what the frame shows; at worst it asks for something slightly different, and the model drifts away from the frame to match the words.

## Step 2: Write the shot in six parts

Write the shot in six parts, in this order:

1. **Start state:** where she is when the clip begins. She stands still, looking at the chameleon.
2. **One main action:** she picks up the lemon and holds it out.
3. **End state:** where the action lands. She ends holding the lemon up, smiling. Without it, the model may repeat the action, or add a new one.
4. **Camera:** the shot size, and one move. Here a medium-long shot at table height, with a slow push-in.
5. **Pace:** gentle, real-time.
6. **Sound:** what you hear in the room. Write "no music" if you'll add your own later.

The prompt the film uses:

```text
The girl with the long white bubble braid stands still beside the kitchen table, looking at her green chameleon on the edge of the fruit bowl. She steps up to the table, picks up the lemon with both hands and holds it out to the chameleon, which slowly turns its head and both eyes toward the lemon. She ends standing at the table, holding the lemon up in front of the chameleon, smiling. Medium-long shot at table height, slow push-in toward the girl and the chameleon. Gentle, real-time pace. Sound: soft kitchen room tone, birds outside the window, light footsteps on the tiles. No music, no dialogue.
```

![The end of the six-part prompt in the Prompt field: the camera, the pace and the sound](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/2fe65bbfc2f488a2cfecba97f48ee5ffb200bfc2317a598f5cc80d2bc2fc3e74.webp)

Name the camera move exactly ("slow push-in", "slow arc to the right") rather than with a vague word like "cinematic". Describe motion over time, not a frozen picture.

## Step 3: Set the duration and generate

Open **Settings** and choose a **Duration** that fits the action. If it's too long, the model fills the time with extra moves. One action with a clear end state needs only a few seconds; the film uses **8s**.

The other settings in the film:

- **Resolution:** **480p**. While **Draft** is on, the output is always 480p, whatever this says.
- **Aspect Ratio:** **Auto**. With a first frame, the clip keeps the shape of the image.
- **Generate Audio:** on, so the model makes the sound the prompt describes.
- **Output Format:** **MP4**. MOV keeps more color for grading, in larger files.
- **Draft:** on, for a quick 480p preview.

![Settings for the six-part shot: Duration 8s, Resolution 480p, Aspect Ratio Auto, Generate Audio on, Output Format MP4, Draft on](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/ed4b3c3603d9cb85b872673c8f36dfc74f7773b567f10369b272fa853f8ed50b.webp)

The price shows on the **Generate** button. Click **Generate**.

She walks to the table, picks up the lemon, and stops on the end state, while the camera pushes in.

![The six-part result: the girl holds the lemon out to the chameleon. The banner reads "480p draft, 6 days left to finalize in 1080p"](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/36c05e5906352a1b946c0d9912d045a239a9be7e569bfbd501fba45c69944740.webp)

The banner above a draft shows how long you have left to finalize it. Click **Finalize in 1080p** to open **Seedance 2.5 Draft Complete** with the draft already loaded, then click **Generate** for the 1080p version. A draft can be finalized for 7 days.

## Step 4: Time the beats of a longer shot

For a longer shot, split the action into beats, and give each beat its time. One action per beat: the lemon, the tongue, the color change, and the smile. Start each beat on its own line with its time in brackets, for example `[0-4s]`.

Keep one camera move for the whole shot. Here, a slow arc around the table.

The prompt in the film, as far as its sound line (the end of the prompt is not shown in the film):

```text
One continuous shot, no cuts. The girl with the long white bubble braid stands beside the kitchen table, looking at her green chameleon on the edge of the fruit bowl.
[0-4s] She steps up to the table, picks up the lemon with both hands and holds it out to the chameleon.
[4-8s] The chameleon shoots out its long sticky tongue, grabs a green grape from the bowl and pulls it into its mouth. She laughs, surprised.
[8-12s] She sets the lemon down right next to the chameleon. The chameleon's skin slowly changes color from green to the same bright lemon yellow, starting at its head.
[12-15s] She leans in with her chin on her hands, smiling at the yellow chameleon, and it blinks at her.
Camera: medium-long shot at table height. The camera slowly arcs to the right around the table and moves a little closer during the whole shot, ending on a medium shot of the girl and the chameleon.
Pace: gentle, real-time.
Sound: soft kitchen room tone, birds outside the window, a tiny wet snap when the tongue catches the grape.
```

![The timed-beats prompt in the Prompt field, from "One continuous shot, no cuts" to the [4-8s] beat](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/31489186865b0ac18efd15426fa11eef5e2735c07138e4dbfa2222cda01e1aab.webp)

Set the duration to the end of the last beat: here **15s**. Click **Generate**.

![Duration set to 15s in Settings](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/5766b014259404d5df8067f0508e462524cf503041109b554d0df08629db92ee.webp)

Watch the result beat by beat: each action should land in its own window while the camera keeps arcing.

![The timed-beats result at the last beat: the chameleon has turned lemon yellow and the girl leans in, smiling](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/8f404ab76c92a39747e80dc0f22e583dc2f210750dfec6ad01b06ecb7d39b339.webp)

## Step 5: Ask for cuts

You can also ask for cuts, so one clip holds several shots:

- Write each shot on its own line, with its time, its framing and its camera.
- Start each new shot with "Cut to".
- Say that the girl, the chameleon and the kitchen stay the same in every shot. Otherwise the model may treat each shot as a new scene.

The film's clip has three shots in 12 seconds: a medium-long static shot (0 to 4 s, she picks up the lemon), an extreme close-up (4 to 8 s, the chameleon turns yellow) and a close-up with a slow push-in (8 to 12 s, her smile). Its prompt opens like this:

```text
Three shots with hard cuts. The same girl, the same chameleon, the same kitchen and the same sunlight in every shot.
Shot 1 [0-4s]: medium-long shot at table height, static camera. The girl with the long white bubble braid stands beside the table, then steps up and picks up the lemon from the fruit bowl.
Cut to shot 2 [4-8s]: extreme close-up of the green chameleon on the edge
```

The film shows only this part of the prompt. Shot 3 follows the same pattern, starting with "Cut to shot 3".

![The cuts prompt in the Prompt field: "Three shots with hard cuts", Shot 1 and the start of shot 2](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/f08b88719e222fc4367d2604136b6f72e7f3c60f4f3505b1d96293cf0d1b339e.webp)

Set the duration to the end of the last shot, and click **Generate**.

![Shot 2 of the cuts result: an extreme close-up of the chameleon, now yellow, looking at the lemon](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/5531956eabe415aee97d2bfb961e66f8069e898475bf813fc9d386702f87634b.webp)

:::check
**Accept when:** the clip starts on the first frame, every action you wrote happens once and in order, it ends on the end state you described, the camera makes the one move you asked for, and the girl, the chameleon and the kitchen stay the same from start to end (and across cuts).
**Reject when:** an action repeats or a new one appears, the clip ends mid-action, the camera does something else, a beat is skipped, or a cut brings in a different character or room. Fix the prompt, or the duration, and run another draft.
:::

## What to avoid

- **A prompt that only describes the scene** gives a clip where almost nothing happens. Use active verbs, and say what moves.
- **Several camera moves in one clip** blend into one. For several moves, make several shots.
- **Writing what should not happen.** Write what should happen, not what shouldn't: "doesn't turn around" can make her turn. Write "She keeps facing the table." instead.

## Sound in the prompt

With **Generate Audio** on, the model makes the sound from your words. Describe the room tone, the small sounds tied to the action, and say plainly what you don't want: "No music, no dialogue."

Seedance 2.5 also reads the developer's bracket channels for sound: ( ) for music and ambience, < > for sound effects, { } for spoken dialogue, and 【 】 for on-screen subtitles. For dialogue, name the language first. The Knowledge Center recommends keeping **Generate Audio** off and adding your own cleared track when the sound matters: the generated soundtrack is checked by moderation after the video is made, so a flagged track can block a finished clip.

## The shape of each prompt

| Shot | What to write | Duration |
|---|---|---|
| One shot | Start state, one action, end state, camera, pace, sound | Just long enough for the action |
| Timed beats | One opening line, then `[0-4s]`, `[4-8s]`... one action each, then one camera line, pace and sound | The end of the last beat |
| Cuts | "Shots with hard cuts", what stays the same, then one line per shot with its time, framing and camera, each new one starting "Cut to" | The end of the last shot |

## Common problems

| What you see | What to do |
|---|---|
| Almost nothing happens | Add one clear action with an active verb, and an end state |
| The action repeats or a new one appears | Add an end state, or shorten **Duration** |
| The clip ends before the action is done | Lengthen **Duration**, or split the action into timed beats |
| Beats run into each other | Give each beat one action only, and set the duration to the end of the last beat |
| The camera wanders or mixes moves | Keep one camera move per shot; for more moves, ask for cuts |
| The look drifts from the first frame | Stop describing the look; describe only what changes |
| A cut brings in a new character or room | Say that the characters, the place and the light stay the same in every shot |
| The model does what you told it not to | Rewrite as what should happen: "She keeps facing the table" |
| A finished clip is blocked | Turn **Generate Audio** off and add your own sound later |

## What's next

To move the camera with more control, read [Direct the camera and motion](https://learn.scenario.com/video/direct-the-camera-and-motion.md). To fix both the start and the end of a shot, use a first and a last frame in [Generate video](https://learn.scenario.com/video/generate-video.md), and to keep a character the same across clips, see [Reference to video](https://learn.scenario.com/video/reference-to-video.md).

Source: https://learn.scenario.com/methods/prompt-video