Cut Editor

A 3-tab system for completing a single cut. Work through Script, Keyframes, and Takes in order to bring each cut to life.

Overview

The Cut Editor is the central editing area that appears when you select a cut from the storyboard. It consists of three tabs -- Script, Keyframes, and Takes -- each representing one stage of cut completion.

+--------------------------------------------------+
|  Properties Bar: Cut # - Status - Thumbnail       |
+----------------+----------------+-----------------+
|  1. Script     |  2. Keyframes  |   3. Takes      |
+----------------+----------------+-----------------+
|                                                   |
|           Content area for selected tab           |
|                                                   |
+---------------------------------------------------+
|  FloatingNavBar: <- Back          Next ->          |
+---------------------------------------------------+

Tab Progression

  • Each tab is labeled with a number: 1. Script, 2. Keyframes, 3. Takes.
  • Completed tabs display a green check icon.
  • Tabs with generation in progress display a loading icon.
  • The Next button in the bottom FloatingNavBar is enabled only when the current step's prerequisites are met:
    • Script tab: A script must be written before Next is enabled
    • Keyframes tab: A representative keyframe must be selected before Next is enabled

Properties Bar

A collapsible bar above the tabs that displays summary information for the current cut:

  • Cut number (e.g., [01])
  • Current status (Pending / Keyframe Generated / Video Generated / Completed)
  • Selected keyframe thumbnail
  • Selected take thumbnail

Tab 1: Script

The Script tab provides a block editor for writing your script. It offers three block types that follow the structure of a film screenplay.

Block Types

BlockIconColorPurposeExample
Location📍Blue left border + blue backgroundSet the scene/location@Cafe - Afternoon
Action👁Neutral left borderDescribe actions, situations, directionsWarm sunlight streams through the window
Dialog💬Purple left border + purple backgroundCharacter dialog"Huh...?"

Adding Blocks

  1. Enter key: Press Enter at the end of a block to automatically insert a new Action block.
  2. + button: Click the + button at the bottom of the editor to choose a block type (Location / Action / Dialog) and add it.
  3. Slash (/) command: Type / in an empty block to open the block type picker. Use arrow keys to navigate and Enter to select.

Switching Block Types

Click the icon on the left side of a block to open a dropdown menu where you can switch to a different block type. The block's text content is preserved.

Location Block

Sets the scene's location.

  • Typing triggers autocomplete with registered locations.
  • Use arrow keys and Enter to select, which inserts the @LocationID.
  • You can also type a location name manually as plain text.

Action Block

Describes actions, situations, and directorial notes.

  • Typing @ triggers a unified @mention autocomplete showing both characters and locations.
  • Characters appear with a person icon and locations with a pin icon.

Dialog Block

Contains a character's spoken lines.

  • A character selection dropdown appears at the top of the block.
  • Select the speaking character from the list of registered characters.
  • Dialog text is displayed in italic style.

Script Example

📍 @Cafe - Afternoon
👁 Warm afternoon sunlight streams through the window. Soft jazz plays.
👁 @Seohee sips her coffee while reading a book, then looks up.
💬 Seohee: "Huh...?"
👁 A man sitting at the table across from her. A familiar silhouette.

Auto-Save

Scripts are auto-saved with a 2-second debounce. There is no need to press a save button. Your work is also auto-saved when switching to another cut.

Tab 2: Keyframes

The Keyframes tab is where you generate AI images (keyframes) based on the script and select a representative image.

Workflow

1. Select an Image Model

Choose an AI image model at the top of the tab. Recommended and compatible models are shown based on your project settings.

See AI Models & Quality for model characteristics and costs.

2. Generate the AI Image Prompt

Click the [Generate Prompt] button and the AI (Gemini) will analyze the script and write an image generation prompt automatically. This step is free.

You can edit the generated prompt directly. Typing @ in the prompt field triggers character/location autocomplete.

3. Reference Images Panel

Characters and locations mentioned with @ in the prompt are displayed automatically.

  • Characters: Thumbnail + variation dropdown. Select a different variation's sheet to generate with a different outfit for this cut only.
  • Locations: Thumbnail + variation dropdown. Select a different version (e.g., day/night) of the same location.
  • Additional references: Choose extra reference images from the asset browser (up to 5).

4. Generate Images

Set the number of images to generate (1--4) and click the [Generate Images] button. A loading animation is displayed during generation.

Once complete, images appear on the keyframe canvas:

  • Click an image to view it in a lightbox.
  • Hover over an image to see the model badge and [Select] button.

5. Select the Representative Keyframe

Click the [Select] button on your preferred image to designate it as the representative keyframe. The selected image gets a blue border and checkmark, and its thumbnail is reflected in the cut list on the left.

Tip: If none of the images are satisfactory, modify the prompt or switch models and regenerate. Previously generated images are preserved.

Uploading Images

You can also use your own images as keyframes. Click the [Upload] button at the top of the canvas to add images.

Tab 3: Takes

The Takes tab is where you generate AI videos (takes) based on the selected keyframe and choose a final take.

Workflow

1. Select Generation Mode

ModeDescriptionBest For
KeyframeUses the selected keyframe as the first frame of the videoWhen you need visual continuity from the keyframe
ReferenceUses the keyframe + Character Sheet as references for video generationWhen you want stronger character consistency

2. Video Model & Audio

Select a video model, and enable the audio toggle if you need sound.

Models with audio support: Seedance 1.5 Pro, LTX 2 Fast, Kling 2.6 Pro, and others. See Video Generation for details.

3. Generate the AI Video Prompt

Click [Generate Prompt] to auto-generate a video prompt from the script and keyframe information (free).

4. Camera Motion

Choose from 19 camera motion options:

CategoryMotions
BasicStatic
HorizontalPan Left, Pan Right
VerticalTilt Up, Tilt Down
ZoomZoom In, Zoom Out
DollyDolly Forward, Dolly Back
TrackingTracking, Orbit
CraneCrane Up, Crane Down
EmphasisPush In, Pull Out, Dolly Zoom
CinematicReveal, Whip Pan, Handheld

5. Negative Prompt (optional)

Enter text describing elements you want excluded from the result. Supported by some models (Wan, Veo, Kling).

6. Duration

Select the video duration. Available options vary by model (typically 4--20 seconds).

7. Generate Video

Click the [Generate Video] button. Video generation runs asynchronously and typically takes 2--6 minutes.

8. Review and Select a Take

Review the generated videos in the take grid:

  • Processing: A progress bar and estimated time are shown.
  • Complete: Hover to auto-play the video. Click for playback controls.
  • Failed: An error message is shown along with a [Retry Check] button.

Click the [Select] button on your preferred take to designate it as the representative take. The selected take gets a blue border and checkmark.

Tip: If you are not satisfied, change the prompt, camera motion, or model and generate additional takes. Previous takes are preserved.

Step-by-Step Guide: Completing a Cut

1. Write the Script (Script Tab)

  1. Use a Location block to set the scene (e.g., @Cafe - Afternoon).
  2. Use Action blocks to describe the situation.
  3. Use Dialog blocks to write character lines.
  4. Click Next to move to the Keyframes tab.

2. Generate Keyframes (Keyframes Tab)

  1. Click [Generate Prompt] to create an AI image prompt.
  2. Edit the prompt if needed.
  3. Check the character/location variations in Reference Images.
  4. Click [Generate Images] to produce 4 candidate images.
  5. Select the best image as the representative keyframe.
  6. Click Next to move to the Takes tab.

3. Generate Takes (Takes Tab)

  1. Choose a generation mode (Keyframe recommended).
  2. Click [Generate Prompt] to create an AI video prompt.
  3. Set camera motion and duration.
  4. Click [Generate Video].
  5. Review the generated takes and select a representative take.

Repeat this process for every cut to complete your short-form drama. See Storyboard & Cuts for overall storyboard management.

Related Docs