​

Script Generation

The Script Generation node uses AI to break a video concept into a structured multi-shot script. Each shot is a self-contained scene description optimized for AI image and video generation.

​

Inputs

​

Outputs

​

Available engines

​

How it works

The AI takes your video concept and breaks it into a specified number of shots (configured in the flow). Each shot description includes:

• Subject — Who or what is in the scene
• Action — What’s happening
• Environment — Where the scene takes place
• Lighting — Visual mood and light direction
• Camera angle — Perspective and framing

Each shot is 15–40 words of cinematic, self-contained scene description. The output is a numbered list with no preamble.

​

Configuration options

​

Credit cost

1 credit per generation (Gemini Flash).

​

Tips

• Write clear, specific video concepts — the more detail you give, the better the shots
• The generated shots are optimized for AI image/video generation downstream
• Each shot is designed to be self-contained — it can be used independently for image or video generation
• Connect this to a Parse Scenes → Split → Image/Video Gen → Collect pipeline for multi-shot video creation

​

Example use cases

• Breaking “A day in ancient Rome” into 5 cinematic shots
• Creating a 3-shot product showcase sequence
• Generating a 7-shot storytelling narrative