Skip to content

JSON Prompt Generator for AI Video and Image

Fill in the subject, camera, lighting, and style, and watch a clean, indented JSON prompt build itself for Sora, Veo, Kling, and Runway.

Runs in your browser · nothing uploaded

Start from a preset

Your JSON prompt

Fill in a field or two and your structured JSON prompt appears here, ready to paste.

Modern AI video and image models follow structured input far more reliably than a run-on sentence. This json prompt generator for ai video turns plain fields like subject, action, setting, camera shot, lighting, style, color palette, aspect ratio, and duration into a tidy JSON object you can paste straight into your model of choice.

It works as a json prompt builder for both motion and stills. Pick video for a json prompt for Sora, Veo, or Kling with a duration field, or switch to image for an ai image json prompt. Only the fields you fill in appear in the output, so you never ship empty keys, and everything happens in your browser.

How it works

  1. 1

    Choose video or image

    Set the output type. Video adds a duration field for a structured ai video prompt, image drops it for a still. You can also load a Cinematic or Product shot preset to start fast.

  2. 2

    Describe the shot

    Type the subject and action, then set the scene, camera shot type and movement, lighting, style, color palette, and aspect ratio. Leave anything blank and it simply stays out of the JSON.

  3. 3

    Copy or download the JSON

    The indented json prompt updates live as you type. Copy it to the clipboard or download a .json file, then paste it into Sora, Veo, Kling, Runway, or an image model.

Questions about JSON prompts for AI video

What is a json prompt for Sora, Veo, or Kling?
It is a prompt written as a JSON object with named fields like subject, camera, lighting, and duration instead of one long sentence. Structured keys give the model clearer instructions, so shots come out closer to what you pictured. This tool assembles that object for you and only includes the keys you fill in.
How do I write a structured ai video prompt?
Start with the subject and the action, then add the setting, camera shot type and movement, lighting, style or mood, color palette, aspect ratio, and duration in seconds. Keep each value short and concrete. The json prompt builder here lays those out in the right shape automatically.
Can I use this as an ai image json prompt builder?
Yes. Switch the output type to image and the duration field disappears, leaving a JSON object suited to image models. The subject, camera, lighting, style, palette, and aspect ratio fields all carry over.
Is it free?
Yes, it is completely free with no sign up. Nothing you type is uploaded, since the JSON is built entirely in your browser.

Building something bigger?

Hivly is made by CodingEagles, a software studio that ships production web apps. If you have a real project, get in touch.

See what CodingEagles does →