UX UI prompt generator Create a UI design prompt for any AI UI design generator
This free UX UI prompt generator turns what a screen or flow needs to solve into a structured text prompt for ChatGPT, Claude, Gemini, Google Stitch, v0, Lovable or any AI UI generator. It’s built for UI designers, product designers and product teams. Founderz made it; our Certificate Program in Artificial Intelligence and Innovation teaches people to use AI at work.
- Free
- About 2 minutes
- Runs in your browser
- ChatGPT, Claude, Gemini and Copilot
At a glance
- What it does: Turns your brief into a structured UX/UI design prompt.
- Which tasks: Screens, flows, specs, audits and prompts for Figma or v0.
- What you get: a prompt ready to copy, plus follow-up requests.
UX/UI prompt generator
Describe the screen or flow you need to solve and get a prompt that makes ChatGPT, Claude or Gemini deliver screens, flows, specs or a prompt ready for Figma, Stitch, v0 or Lovable. Only step 1 is required: everything else stays on Auto.
How to use the UX UI prompt generator
Only the first step is required: anything you skip stays on Auto.
- The essentials. Write what the experience needs to solve, pick what you need (from one key screen to an audit or a prompt for a UI tool) and, if you like, turn on “Have the AI ask me questions before designing”.
- Screen and user. Pick the screen or flow, the primary user and the platform. “Advanced settings” holds “Content and data”, where you choose realistic copy, fictional data or structure only.
- Design and states. Set the visual direction, the density and the states to cover. In “Advanced settings” you can fix the priority components and “System and responsive”.
- Tool and delivery. Choose the design tool (Figma, Stitch, v0, Lovable, Uizard or an image generator) and the interface language. The advanced settings add the material you provide, extra controls such as acceptance criteria or a user test plan, and your constraints.
- Generate and copy your prompt. Click “Generate prompt”, enter your email to see the whole prompt and copy it.
Anything you leave on Auto, the AI infers from your brief and records as an assumption. If you’ll attach a screenshot or wireframe, pick it under “Material you will provide” in “Advanced settings” so the model works from what it can see.
Example: a UX UI prompt to audit a mobile checkout
With the generator’s “Checkout” example, the prompt starts like this. It’s an excerpt: the button copies the full prompt.
ROLE Act as a senior UX/UI designer with digital product experience. You start from the user and their task, not from aesthetics: you apply visual hierarchy, Nielsen's usability heuristics, well known interface patterns and WCAG 2.2 AA, and you never fill gaps with invented user data. GOAL Audit the current experience and deliver a list of improvements prioritized by impact and effort. CONTEXT Design brief provided by the user (treat it as your main source): [START OF BRIEF] Online store for climbing gear. Checkout has 4 steps, forces account creation before payment and on mobile the keyboard hides the continue button. I want to know what to change first. [END OF BRIEF] ... OUTPUT FORMAT Deliver a findings table with columns No., Where, Issue, Heuristic, Severity (0 to 4), Proposed fix, Impact and Effort; then the 3 quick wins and the 3 structural changes. ...
The full prompt adds the WCAG 2.2 AA accessibility constraints, seven working steps, the exact findings table format with a sample row and four final checks.
The result: instead of a vague list of tips, the AI returns a findings table prioritized by impact and effort, with a concrete fix for each issue.
Tips: the best AI results for app design
6 quick tips
- Describe the user’s goal in the brief with a verb and a context (“pay in under a minute on a phone”) instead of asking for “a nice screen”. Clear user goals lead to better design decisions.
- Choose “Empty, loading, success, error” under “States to cover”: they’re the states generated screens tend to skip.
- Check the contrast of whatever the tool returns: WCAG 2.2 asks for a ratio of at least 4.5:1 for normal text.
- If you have a design library, choose “Tokens from an existing system” under “System and responsive” and name your components in “Context, constraints or references (optional)”, so the output stays customizable.
- Iterate one variable at a time (the user, the density or the visual direction) and compare the results side by side.
- Treat the output as a first draft and run a quick usability test before it goes to development.
Use cases: from text prompt to multi-screen flow
- Onboarding A short flow for a new user who needs guidance, with a step-by-step progress pattern.
- Data dashboards A dashboard for an internal team that keeps dense analytics readable.
- Handoff A spec of components, interactions and states developers can build from without extra meetings.
- Audit A prioritized list of fixes for a screen you already have, based on a screenshot.
What makes good UX design prompts
A good prompt covers six things: the role the model takes, the user’s task, the context (platform, device, moment of use), the material you provide, the constraints (accessibility, privacy, design system) and the output format. Without the task and the user persona, a design AI fills the gap with generic interaction patterns.
With them, it can reason about hierarchy, navigation and microcopy at each step, and you can judge each proposal against real user needs. The same structure works for websites and mobile.
More AI prompts: design prompts for ChatGPT and other tools
If you’d rather start from finished examples, the Founderz prompt library has ready-made examples for design, product and other tasks. If you use Midjourney for visual moodboards, the image tool below fits better.
Why I built these generators
Frequently asked questions
Is this tool free?
Yes, it’s free. Before it asks for anything, it shows you a preview with the first sections of the prompt. To see the whole prompt and copy it, enter your email and accept the Privacy Policy; receiving communications from Founderz is optional.
It only asks once per visit: any prompts you generate after that are shown in full.
Does it generate UI or create UI designs for me?
No. It writes the prompt; the screens come from the tool you paste it into, such as v0, Lovable or Google Stitch, where the model generates UI from simple text. Your design judgment still decides which proposal works.
Can I use it with screenshots?
Yes. Choose “Screenshot of the current screen” under “Material you will provide” and attach the file in your tool. The prompt tells the model to describe what it sees first and point out each issue with its location before proposing changes.
Does it work for UX research?
It helps you prepare: under “What do you need?” you can ask for an audit with prioritized improvements, and under “Extra controls” you can add a “User test plan”. Interviews and usability sessions with real users still come from you.
What language does the prompt come out in?
In the language of your brief, unless you pick another under “Interface language”. There you can ask for the interface copy in English, in Spanish or in both in parallel columns.
Certificate Program in Artificial Intelligence and Innovation
If you want to move from one-off prompts to an AI-driven workflow across your team, the Founderz Certificate Program in Artificial Intelligence and Innovation teaches you how. It is fully online and developed in collaboration with Microsoft.
Last updated:
