●Last checked 14 prompts1 of 14 run in Lovable

    Lovable UI prompts: 14 for screens, empty states and dashboards

    Prompt builder · five steps · no account

    Step 1 of 5

    What are you building?

    No account needed.

    Building something else? Describe it to the AI generator

    Lovable builds the happy path. Empty, loading, error and mobile states only show up when a prompt names them. These 14 prompts name them, and this page tells you which one to run first, what it cost us and where it goes wrong.

    By Marco Kohns · Prompts written and run by us

    Short answers

    Short answers

    Where do I start?

    Navbar, then header and footer, then empty states and loading skeletons. Every screen you build after that inherits them.

    Pick by situation →

    What does a run cost?

    9.5 credits

    Our measured run of the states prompt, Cloud setup included. On the Free plan that is two days of daily credits.

    See the credit math →

    Will Lovable ask design questions?

    Not for dashboards or admin panels: they build straight away. Visual pages like landing pages get up to four questions. Neither costs extra credits.

    How to skip them →

    My app already exists

    Run empty states and loading skeletons, then the modal and dialog system. Both standardise screens you already have instead of adding new ones.

    Fix lines for live apps →

    14

    UI/UX prompts in this category

    ↳ Copy-paste, free

    9.5credits

    Measured run, empty states & loading skeletons

    ↳ 77.3 → 72.8 after 5 daily credits

    2

    Questions Lovable asked first

    ↳ Before writing any code

    37

    Upvotes since Jan 2026

    ↳ One per visitor, self-reported

    Data basis: one run of one prompt in a fresh Lovable project, balance read before and after. The other 13 prompts are not run yet and say so on their card. Credit rules from Lovable's docs, checked 22 October 2026.

    Find a prompt

    Which screen are you building?

    Cards are sorted by build order. Each says whether we ran it in Lovable, and if not, shows a question mark instead of a guess.

    Showing 14 prompts

    • Step 1 · Frame▲ 2

      Responsive Navbar Builder

      A server-rendered responsive navbar from one config module: a mobile Sheet below 768px, active states shown by underline and weight, and full keyboard and screen-reader support.

      Run it when: first, in a new project, before any feature screen.

      Not run yet · credits ?
      Try in Lovable
      Pro version adds the growth-channel instructionsUnlock
    • Step 2 · States▲ 1

      Empty States & Loading Skeletons

      Reusable empty, loading, error and partial-data components with rules for when each appears, plus a gallery page that renders every variant for review.

      Run it when: right after the app shell, before feature screens multiply.

      Measured: 9.5 credits · asked 2 questions first
      Try in Lovable
      Pro version adds the growth-channel instructionsUnlock
    • Step 3 · Dialogs▲ 1

      Modal & Dialog Component System

      One dialog system for the app: a base modal in 3 sizes and a confirm dialog with shared focus, dismissal and unsaved-input rules, plus bottom sheets on mobile.

      Run it when: before your first confirm, edit or delete dialog.

      Not run yet · credits ?
      Try in Lovable
      Pro version adds the growth-channel instructionsUnlock
    • Step 4 · Data screens▲ 1

      Table UI with Sorting & Filters

      A data table where sorting, filtering and pagination run in the database through one allowlisted server function, the view lives in the URL, and export follows the filters.

      Run it when: any list longer than one screen, or one users need to filter.

      Not run yet · credits ?
      Try in Lovable
      Pro version adds the growth-channel instructionsUnlock
    • Step 4 · Data screens▲ 6

      Analytics Dashboard UI

      An operator dashboard with four metric tiles, a trend chart and a source table under one date range, where a failed query shows a dash with Retry.

      Run it when: once you have data to show, and someone who decides from it.

      Not run yet · credits ?
      Try in Lovable
      Pro version adds the growth-channel instructionsUnlock
    • Step 4 · Data screens▲ 18

      File Upload + Tagging UI

      Drag-and-drop uploads to a private per-user bucket with five named file states, server-side type and size checks, and normalised tags for filtering the library.

      Run it when: users bring their own files into the app.

      Not run yet · credits ?
      Try in Lovable
      Pro version adds the growth-channel instructionsUnlock
    • Step 4 · Data screens

      Location-Based Listings UI

      A location directory with a server-rendered page per place and per city, plus a synced map and list browse layer on top for filtering.

      Run it when: each place or city should get its own page in search.

      Not run yet · credits ?
      Try in Lovable
      Pro version adds the growth-channel instructionsUnlock
    • After step 4

      Settings & Preferences Panel UI

      A settings area with deep-linked sections, auto-save per control that restores the old value on failure, a notification grid and a guarded Danger zone.

      Run it when: accounts exist and users need to change something.

      Not run yet · credits ?
      Try in Lovable
      Pro version adds the growth-channel instructionsUnlock
    • After step 4▲ 2

      User Onboarding Flow UI

      A 4-step onboarding flow with progress stored on the server per user, resumable across devices, skippable steps recorded as skipped and a review step.

      Run it when: setup takes more than one screen.

      Not run yet · credits ?
      Try in Lovable
      Pro version adds the growth-channel instructionsUnlock
    • Standalone

      Poll Tool UI

      Single-question polls with one vote per visitor enforced by a unique constraint on the server, results showing the vote count beside percentages, and no reset.

      Run it when: you want a small first project to learn Lovable on.

      Not run yet · credits ?
      Try in Lovable
      Pro version adds the growth-channel instructionsUnlock
    • Before launch▲ 1

      Simple 404 & Error Screens

      A server-rendered 404 that returns status 404 with path-aware search, a per-page error boundary that keeps typed input, an offline banner and server-side logging of every error.

      Run it when: before you share a link anywhere.

      Not run yet · credits ?
      Try in Lovable
      Pro version adds the growth-channel instructionsUnlock
    • Before launch▲ 1

      SEO Meta Tag Generator UI

      An internal editor for page titles and meta descriptions: sitemap import, a live search-result preview, pixel-width clipping and duplicate flags, revision history and CSV or JSON export.

      Run it when: before you publish pages you want found in search.

      Not run yet · credits ?
      Try in Lovable
      Pro version adds the growth-channel instructionsUnlock
    • Before launch▲ 3

      Social Share Widgets

      Sharing for results and articles: server-rendered Open Graph metadata per URL, native share on phones, platform links and copy-link on desktop, and no third-party scripts.

      Run it when: users should share pages from your app.

      Not run yet · credits ?
      Try in Lovable
      Pro version adds the growth-channel instructionsUnlock

    Building the whole UI? The bundle has all 14 as one kit in build order, with every Pro version unlocked, plus 138 more prompts.

    See what is in the 152-prompt bundle

    Disclosure: if you create a Lovable account through links on this page, we may earn a commission. It costs you nothing and does not change what we recommend.

    Measured run

    We ran the states prompt in a fresh project. This is what happened.

    ●●● prompt · excerpt · Empty States & Loading Skeletons Full prompt on the card above

    ▍Context

    Build the application's states system: reusable empty, loading, error and partial-data components with clear rules for which appears when. Every list, form and detail screen will use these; they are the difference between an app that guides and one that shrugs.

    ▍Core Features

    1. EmptyState component: icon slot, one-line explanation, primary action, sized variants

    2. Skeleton components shaped like the final content: list rows, cards, detail blocks

    3. ErrorState: what failed in plain words, retry where retry can work, reference id where it cannot

    4. Distinct first-use empty (never had data) versus cleared empty (had data, now none)

    ▍State Rules

    - Zero data and failed-to-load are never the same screen; conflating them hides outages

    - Skeletons match final layout dimensions so completion does not reflow

    - Every empty state names the action that fills it; a dead-end empty is a design bug

    ▍Before Building

    Ask me any questions you need in order to fully understand the audience and the screens, then save the decisions to the project knowledge so later prompts build on them.

    ▍Distribution hooks · Pro

    Locked. The Pro version adds the growth-channel instructions for this screen.

    1. 01

      Pasted into a fresh project

      No template, no design system, default settings. Visual Direction line removed, as the prompt suggests.

    2. 02

      Lovable asked 2 questions before writing code

      Who will use the app, and which screen should get the states first.

    3. 03

      We answered: a small task tracker

      Lovable switched on Lovable Cloud for task data and sign-in.

    4. 04

      It built 4 base components, a /states gallery and a working task list

      Every state can be triggered from the gallery.

    5. 05

      It tested the states itself

      Creation, completion, slow loading, failure and retry, then removed its test data.

    6. 06

      Balance read: 77.3 to 72.8, after the day's 5

      9.5 credits for the whole run, Cloud provisioning and sign-in included.

    Lovable project chat right after the states prompt was sent, showing a question card that reads Who will use this app, and what kind of information will its screens show, with an answer field, Skip all and Next
    Questions before code
    The generated states gallery page in Lovable's preview: an Empty section with two cards, First use reading No tasks yet with an Add your first task action, and Cleared / filtered reading No matching tasks with a Clear filters button
    The gallery Lovable built
    The Error section of the generated gallery: a retryable card with a Try again button, a non-retryable card with a reference ID, and a compact form failure card
    Retryable vs non-retryable
    Lovable plans and credit usage page after the run: 72.8 credits left, daily build credits 0 left, and a usage list showing the State Guider project at 9.50 credits
    Balance after the run

    Credit math

    What 9.5 credits means on your plan

    Lovable prices build messages by how complex the request is and how much work gets done. Treat our number as one data point, never as a price list.

    Where a 9.5-credit run comes from on each Lovable plan
    PlanDaily build creditsOne states run (9.5)Unused credits
    Free5 a day, max 30 a monthTwo days of daily credits. Three runs this size fit in one month's cap.Daily credits do not roll over
    Pro5 a day, plus 100 monthly plan credits from €255 from the day's credits, 4.5 from plan credits (what happened in our run)Plan credits roll over, expire after 2 months
    Business5 a day, plus 100 monthly plan credits from €50Same split as ProSame as Pro

    Lovable's own examples, for scale

    “Make the button gray”
    0.50
    “Remove the footer”
    0.90
    “Add authentication”
    1.20
    Build a landing page with images
    2.00

    Source: Lovable docs, Credits and usage, marked illustrative there. Checked 22 October 2026.

    Three ways to spend less

    • Run the big prompts on a fresh day. Daily credits are spent first, so a 9.5-credit run costs you only 4.5 plan credits.
    • Answer the clarifying questions fully. A wrong guess costs a second build message; an answer costs nothing.
    • Build the frame and states before screens. Retrofitting states into 10 screens is 10 messages. Doing it first is one.

    Start here

    Which UI prompt should you run first?

    Pick the sentence that sounds like your project.

    Build the frame first, then the screens

    1. 01Responsive Navbar Builder · sets navigation and breakpoints
    2. 02Custom Footer & Header Templates · one shell for every page
    3. 03Empty States & Loading Skeletons · the states every screen reuses
    4. 04Modal & Dialog Component System · one way to confirm, edit and delete

    Build order

    Four steps. Each one inherits the last.

    1. 01

      Frame

      Navbar · Header & footer

      Navigation, breakpoints and the page shell.

    2. 02

      States

      Empty states & skeletons

      What every screen shows when data is missing, slow or broken.

    3. 03

      Dialogs

      Modal & dialog system

      One way to confirm, edit and delete.

    4. 04

      Data screens

      Table · Dashboard · Upload · Listings

      The screens users came for, inheriting all of the above.

    Anatomy

    What each section of a UI prompt does

    What each prompt section tells Lovable and what you get without it
    SectionWhat it tells LovableWhat you get without it
    ContextWho uses the screen and what data it showsSample data and generic copy
    Core featuresNumbered components, by nameA screenshot instead of an interface
    State rulesEmpty, loading, error and partial dataThe happy path only
    Visual directionFonts, colours, a reference or a design systemDesign questions or Lovable defaults
    Build orderWhat to build firstEverything at once, harder to fix
    Safe-guardsWhat not to touchRegressions in screens that worked
    Before buildingAsk questions firstGuesses where your context is missing

    A one-line prompt gets you

    • Sample data and generic copy
    • A screenshot instead of an interface
    • The happy path only
    • Design questions, or defaults

    A prompt from this page gets you

    • Named components with named states
    • Empty, loading, error and partial data handled
    • A committed visual direction
    • Questions first, then the build

    Fixes

    Where UI prompts go wrong, and the line that fixes it

    Already built something that looks off? Paste the fix line into your existing project.

    Symptom

    Every screen has its own spinner

    Cause: the app was prompted screen by screen, so each one invented its own loading and empty treatment.

    Fix line

    Use the shared EmptyState, Skeleton and ErrorState components from /states on this screen. Do not create new ones.

    Symptom

    An empty list shows nothing

    Cause: no empty state was named, so Lovable rendered an empty container.

    Fix line

    Add a first-use empty state with one sentence and one action, and a separate cleared-filters state with a Clear filters button.

    Symptom

    Looks right in a screenshot, breaks on a phone

    Cause: the prompt described a look instead of naming components and breakpoints.

    Fix line

    Use shadcn/ui components by name. Breakpoints at 640, 768 and 1024 px. Collapse the navbar at 768 and test at 390 px before reporting done.

    Symptom

    Lovable guessed wrong

    Cause: the prompt skipped questions, so Lovable filled your context with assumptions.

    Fix line

    Before building, ask me up to 3 questions about who uses this and what data it shows. Wait for my answers.

    Upvotes

    Which UI/UX prompts do builders upvote?

    Upvotes since January 2026, one per visitor. Small numbers: read them as a hint, never as a ranking.

    • File Upload + Tagging UI18
    • Analytics Dashboard UI6
    • Social Share Widgets3
    • User Onboarding Flow UI2
    • Responsive Navbar Builder2
    • SEO Meta Tag Generator UI1

    The Complete Lovable Bundle

    All 14 UI prompts in build order, plus every other prompt

    Every prompt and kit, every Pro version unlocked. One payment, no subscription.

    One payment. No subscription, no renewal, ever.

    Or get free access first

    FAQ

    Questions about UI/UX prompts

    Can Lovable build a full UI from one prompt?+

    It builds one screen well. A full interface needs the frame, the shared states and the screens prompted separately, in that order. Source: Lovable docs, Prompting best practices.

    Does Lovable show design options before it builds?+

    Yes, by default. Lovable's design guidance shows three directions for a UI request unless the prompt commits to a visual direction by naming fonts, colours or branding, referencing another product or URL, or using a named design system. Every prompt here has a Visual Direction line: fill it to skip the step, leave it out to pick from three.

    How many credits will a UI prompt use?+

    Lovable publishes no per-prompt figure and bills Build mode on the work done. Our one measured UI run, the states system, cost 9.5 credits including Cloud provisioning and Lovable testing its own output. Lovable's illustrative examples run from 0.5 credits for a button colour to 2 for a landing page with images.

    Do the daily credits run out on the Free plan?+

    Yes. Free workspaces get 5 build credits a day, up to 30 a month, and unused daily credits are gone when the day ends. A 9.5-credit run spans two days on Free; three runs that size use up a month.

    Do these prompts assume shadcn/ui?+

    Yes. They name shadcn/ui components because that is what Lovable scaffolds with by default, and a named component is a design instruction the model can act on; a described one is a guess.

    What about dark mode and accessibility?+

    Both are specified in the prompts up front. The navbar and dialog prompts carry focus, aria and keyboard rules; Lovable's own library has a one-sentence dark mode prompt that asks every page, chart and form to look right in both themes, which pairs with the states prompt here.

    Can I use a Lovable design system with these prompts?+

    Yes. In Lovable a design system is a dedicated project holding a React component library, a token schema and setup instructions, connected to other projects in the workspace, at most one per project. Connecting one counts as a committed visual direction, so the three-direction step is skipped.

    What we have not tested yet

    • The other 13 prompts. Their cards show a question mark until we run them.
    • The same screen built with and without named states, side by side.
    • Accessibility audits of the generated screens.

    Next run: Responsive Navbar Builder, the first prompt in the build order. How we test prompts

    Who wrote this

    Marco Kohns wrote and ran these prompts. Lovable Prompts is independent and not affiliated with Lovable.

    “Try in Lovable” links are affiliate links: we may earn a commission when you create an account through them. The Pro versions and the bundle are our own paid products. Neither decides which prompt we tell you to start with.

    ui-ux