Why a short idea needs expanding
The model invents everything you left out: the cat's breed, the time of day, the angle, the drawing manner. Ask for «a cat» and none of ten attempts will match what you pictured — simply because the model has no idea what you pictured.
A full prompt closes those gaps: the subject, what it is doing, where it is, what the light is like and in which manner it is all rendered. The generator does that work for you — it takes the short phrase and writes in the missing blocks.
How it works
Type the idea and press the button. The AI returns a finished description broken into blocks — concept, composition, details, lighting, colours: the very blocks a working prompt is built from, filled in for you.
From there you have two options. Copy the prompt and use it anywhere — it is plain text with no service-specific tags. Or press «Generate from this prompt»: the description lands in the image generator already filled into the field, and you just hit generate.
Writing a prompt costs 1 energy point — roughly ten times less than a generation, so trying different wordings is cheap.
When you don't need a prompt at all
In ready-made tools — greeting cards, caricatures, AI photoshoots, pet portraits — the prompt is built in. There you upload a photo, pick the options, and the service writes the scene description itself.
Prompts matter when you generate an image from scratch or want to depart from the preset: your own scene, an unusual style, something no tool covers.
Frequently asked questions
How much does writing a prompt cost?
1 energy point per attempt — roughly ten times less than generating an image, so trying different wordings stays cheap.
Will the prompt work in other AI tools?
Yes. The generator returns plain text with no service-specific tags or syntax, and every modern image generator understands that kind of description.
Which language should I type my idea in?
Your own. The description is translated and expanded automatically before it reaches the model.
Can I edit the generated prompt?
Yes — it comes back as plain text, so you can edit it before copying or inside the image generator field. You can also write another version of the same idea: every attempt returns a new description.
How is this different from the «Develop idea» button inside a tool?
Same mechanics, but inside a tool the suggestion also accounts for the aspect ratio, style and photos you picked. The separate page is handy when you need the prompt on its own — for another service, say.
