AI prompt generator and builder
Ready-made prompts for photos and photoshoots, chat, study, work, video, code and music — plus a builder that turns one phrase of yours into a full prompt. A 0–100 score shows which meaning-carrying parts are missing: tap any of them right inside the prompt and fill it in. For ChatGPT, Midjourney, Nano Banana, Sora, Suno and others.
How to use it
For example “birthday photo”. Or paste a prompt you already wrote.
The type — image, text, music, video or code — is detected from the phrase.
Missing parts sit right inside the prompt text: tap one and fill it in.
Copy the finished prompt — or run it here and see the answer.
Take a ready prompt for photos, chat or music — or build your own part by part
A pasted prompt is split into parts, and we show what is missing.
What it does
Typical uses
- Write a prompt for a photo session, a portrait, a kids' party or a family shot.
- Take apart a prompt that "somehow does not work" and see what it is missing.
- Take a ready prompt from the library and drop in your own name, age or product.
- Assemble a precise brief for ChatGPT: role, output format and constraints.
- Write a Suno or Sora prompt in English without knowing English.
- Check two versions of a prompt and see the difference in the answers.
Why this one
A typical prompt generator hides the work inside a neural network and hands you a string: why it is better than yours is anyone’s guess, and there is nothing to check it with. Here it is the other way round. The structure is on show: which parts a prompt is made of, which you filled in, which are missing and how much each one weighs. Building, scoring and the library run in the browser. And testing — the thing almost nobody offers: some tools cannot run a prompt at all, others need your own model key. In browsers with a built-in model (Chrome 148, Edge) the test runs on your own device.
FAQ
How is this different from a generator that just rewrites my text?
It shows the structure. A typical generator hands back a finished string and never explains what changed or why. Here you see which parts a prompt consists of, which you filled in, which are missing and how much each affects the result. Someone else’s prompt works exactly the same way: paste it into the same field and the breakdown appears on its own — there is no separate place for it.
Do I need an account?
Building a prompt, the score and the library work straight away — they are computed in the browser. An account is needed only when improve or test go through our neural network. In browsers with a built-in model (Chrome 148, Edge) they run on your own device and need no account either.
Which AI models does it cover?
Text — ChatGPT, Claude, Gemini, DeepSeek, GigaChat, YandexGPT, Grok, Qwen. Images — Nano Banana, GPT Image, Midjourney, FLUX, Stable Diffusion, Ideogram, Kandinsky, Shedevrum. Video — Sora, Veo, Kling, Runway. Music — Suno and Udio.
What is this “model in the browser”, and does everyone have it?
From Chrome 148 and recent Edge the browser carries a language model inside it: it downloads once, lives on disk and computes on your device. The tool probes for it and writes on the button itself which path a run will take. Firefox and Safari have no such model yet — there improve and test go through our neural network, while building, scoring and the library work the same everywhere.
Why are the image prompts written in English?
Because English is the native language of those models and they follow instructions in it more precisely. We measured this on our own music generator: a "no drums" instruction written in Russian was ignored in all six runs, while the English wording was obeyed in all six. So strengthening an image, video or music prompt returns English, while chat and code prompts come back in the language you wrote.
How is the score computed?
From the composition. Each kind of prompt has its own set of meaningful parts with its own weights: for an image, lighting weighs twice as much as palette, because lighting redefines the whole frame while palette only tints it. The score adds up the parts found, subtracts empty words, and cannot go above 40 without the main field.
What does "run the prompt" mean?
The prompt goes to a model and you see its answer. For an image or video prompt there is no generator here, so the model takes the prompt apart in words: what would end up in the frame and which parts stay ambiguous. That is how you catch a mistake before spending a real generation.
Are my prompts saved?
Prompts you build stay in your browser's memory so you can come back to them instead of rebuilding. Clear them with the button in the "Your prompts" section.
Can I use these prompts for commercial work?
The prompts themselves are just text — instructions for a model — and we place no restrictions on using them. What you may do with the RESULT depends on the terms of the model you run it on: Midjourney, Suno, ChatGPT and the rest each set their own rules for commercial use.
We can! Just send us a quick message with your idea. If you'd like to discuss it in detail, leave your email and we'll get back to you. You can stay anonymous.
How do you rate this tool?
Comments
By uploading content you confirm that you are entitled to disclose its contents to third parties, and you undertake not to include other people's personal data in it. See the Terms of Use for details. AI providers Tembrica is an independent product. We are not affiliated with, endorsed by or sponsored by the makers of the models and services listed here; their names are used only to say factually which engine runs a given tool.
