Utilities
img/vid-describe
Any picture or clip, turned into words that describe it.
<!-- Generated by apps/docs/scripts/gen-tools.mjs. Do not edit here: the words live in apps/web/src/lib/tools/describe.tool.json (agent, hint) and packages/shared/src/tools/setting-hints.ts. --> On the rail as **img/vid-describe**. ## When to use it Getting a written account of an image or video: what is actually in it, or the prompt that would recreate it, before reusing or rebuilding the shot. ## When to reach for something else It takes no written direction, so it only reads media you already have. It does not generate or edit an image or video, only the words that describe or could recreate one. ## How to use it 1. Use the card's **Image or video** area. The picture or clip to describe. Drop one from the board or your computer. 2. Pick a mode on the card: **Raw**, **Tripper**, **Dreamer**, **Motion**. 3. Press **Run**. ## On the card | Control | What it does | Choices | | - | - | - | | Image or video | The picture or clip to describe. Drop one from the board or your computer. | | | Run | Starts the run. The number on it is what this run costs in credits, worked out before you press. | | | ×1 | Runs the card several times in a row with the same settings. | 1 to 99 | ## Modes ### Raw What is in the image, in a few sentences you can use as a prompt ### Tripper Same scene, louder — heightened sensory language ### Dreamer Reimagined as if remembered from a dream ### Motion Reads a dropped video's frames and reverse-engineers the Seedance 2 / Kling prompt that would recreate it ## Settings These sit behind the settings button on the card. | Setting | What it does | Choices | | - | - | - | | Model | Which engine answers. The default is a good all-rounder. | Claude Sonnet 5, Gemini 3.5 Flash-Lite, Claude Haiku 4.5, GPT-5.6 Luna |On the rail as img/vid-describe.
When to use it
Section titled “When to use it”Getting a written account of an image or video: what is actually in it, or the prompt that would recreate it, before reusing or rebuilding the shot.
When to reach for something else
Section titled “When to reach for something else”It takes no written direction, so it only reads media you already have. It does not generate or edit an image or video, only the words that describe or could recreate one.
How to use it
Section titled “How to use it”- Use the card’s Image or video area. The picture or clip to describe. Drop one from the board or your computer.
- Pick a mode on the card: Raw, Tripper, Dreamer, Motion.
- Press Run.
On the card
Section titled “On the card”| Control | What it does | Choices |
|---|---|---|
| Image or video | The picture or clip to describe. Drop one from the board or your computer. | |
| Run | Starts the run. The number on it is what this run costs in credits, worked out before you press. | |
| ×1 | Runs the card several times in a row with the same settings. | 1 to 99 |
What is in the image, in a few sentences you can use as a prompt
Tripper
Section titled “Tripper”Same scene, louder — heightened sensory language
Dreamer
Section titled “Dreamer”Reimagined as if remembered from a dream
Motion
Section titled “Motion”Reads a dropped video’s frames and reverse-engineers the Seedance 2 / Kling prompt that would recreate it
Settings
Section titled “Settings”These sit behind the settings button on the card.
| Setting | What it does | Choices |
|---|---|---|
| Model | Which engine answers. The default is a good all-rounder. | Claude Sonnet 5, Gemini 3.5 Flash-Lite, Claude Haiku 4.5, GPT-5.6 Luna |