CommunityImagegithub.com

Store Screenshot Set

Turn real app screens into an App Store or Google Play screenshot set: a short visual story, a poster-like first frame, and frames fitted to the store sizes.

What is Store Screenshot Set?

Makes five to eight marketing screenshots from the app's real UI. The first frame works as a poster, later frames move into the product, and every export matches store dimensions.

Works with~Claude Code✓Codex CLI~Cursor
npx skills add https://github.com/zohaibmushwani/agent-skills/tree/HEAD/skills/store-screenshots

Installed? Explore more Image skills: steipete/songsee, affaan-m/frontend-design, mattpocock/improve-codebase-architecture · View all 6 →

Ask in your favorite AI

Open a new chat with this agent skill pre-loaded.

Documentation

What does Store Screenshot Set do?

Store screenshots are an ad, not a gallery of screens. Visitors decide in one or two seconds, mostly from the first 2-3 frames, so each frame sells one benefit with a big headline and only the slice of UI that proves it. The patterns here come from studying high-performing sets (see references/patterns.md); use them as principles, never copy another app's layout, brand or art.

Inputs (gather first)

  • Real UI screens of the app (PNG). From the app-design-studio skill or captures. Without them, stop: invented UI in store screenshots misrepresents the app.
  • tokens.json or brand colours, logo/wordmark, app name, the one-line promise.
  • Keywords and competitor claims (the aso-research skill, if available) so headlines use the words people search.
  • Platform(s) and languages. Default: Google Play phone 1080 x 1920 and iPhone 6.9" 1290 x 2796.

Workflow (Codex is the renderer; this skill writes its instructions)

  1. Message map. List 5-8 benefits, ordered by what makes someone install. Frame 1 = the promise (poster); frames 2-3 = the core loop; then differentiators, proof, retention. Write story.md: per frame a headline (2-4 words), optional 3-6 word subline, and the real screen that proves it. Rules in references/copy.md.
  2. set.json. Fill the set style once (look, colours with text-on-accent, headline font description, art direction, panorama motif) and the frames (schema in references/frame-recipes.md). The style block is repeated verbatim in every prompt; that is what keeps the set consistent.
  3. Prompts. python scripts/build_prompts.py set.json prompts/ writes NN.md (exact text to send) and NN.refs.txt (real UI screens to attach, plus the approved frame 1 as style reference for frames 2+). Never retype a prompt; send the file.
  4. Three directions (human gate). Generate frame 1 in three set styles (e.g. Poster, Panorama, Character) with Codex - send ONE image first to confirm the image quota (codex_status does not show it). The user picks one; save it as final/01.png (it becomes the style reference).
  5. Generate the rest with generate_image (prompt = file text, reference_images = refs file, background mode for batches). Save as raw/NN.png.
  6. Text and truth check (every frame). Open each image and read it: headline spelled exactly, no extra words, UI not redesigned, no invented logos/ratings. Fix with an edit pass (the frame as reference image + "keep everything identical except: "), max 2 edits per frame; if text still breaks, shorten the headline.
  7. Fit and check. python scripts/fit_store.py raw/ final/play --size play (and --size ios69), then python scripts/check_set.py final/play --set set.json; score the contact sheet with references/rubric.md at thumbnail size. Regenerate the weakest frame first.
  8. Deliver final/<platform>/NN.png, story.md, set.json, prompts/, contact sheet. Label anything made before the app is real as concept.

Non-negotiables

  • UI shown must be the app's real UI (or the approved design marked concept). No fake features, fake ratings, fake awards or "No. 1" claims you cannot prove; social proof only with real numbers the user supplies.
  • No other companies' logos or trademarks unless the user has the right and the store allows it; no real people's likeness from the internet.
  • Every visible word is quoted in the prompt and verified in the output; a frame with a misspelled or extra word is not shipped. Headline must contrast clearly with its background.
  • Keep the first 2 frames understandable on their own, at thumbnail size.
  • Follow store rules: Play screenshots no alpha, 320-3840 px per side, aspect <= 2:1; Apple sizes per device class (re-check official pages).

Files

  • references/patterns.md - what top sets do (8 patterns) and the before/after lessons.
  • references/frame-recipes.md - frame types, layout grid, set.json schema, panorama slicing.
  • references/copy.md - headline formulas, length, tone, localisation.
  • references/rubric.md - 1-second test and scoring.
  • scripts/build_prompts.py (set.json -> Codex prompts + reference lists), scripts/fit_store.py (exact store size, no alpha, contact sheet), scripts/check_set.py; assets/example/ (set.json, prompts, real screens).

Individual skills in this repo

This repo contains 2 individual skills — each has its own dedicated page.

app-design-studio

Turn an app name or a website / web-app idea into a complete high-fidelity UI design repository - pages or screens, every state (loading, empty, error, offline), dialogs, sheets, components, interactions - with a locked design system and measurable usability gates. Three targets, each with its own profile and wording - mobile app (portrait screens: splash, onboarding, auth, home, settings), website (landscape desktop pages plus a narrow responsive layout), web app / portal / dashboard. Use whenever the user gives an app name or idea and wants screens, pages, UI design, a design repo, mockups, a screen inventory, or a design to hand to developers/designers, even if they never say "skill" or "Figma". Also use to audit or fix an existing set of generated screens for usability (contrast, targets, states) while keeping them beautiful.

screenshot-to-compose

Turn app design screenshots (JPG/PNG, often AI-generated) or Figma frames into a native Kotlin Jetpack Compose UI that matches them, design-first. Measures pixels and colours (px to dp), writes screen/asset inventories and a design system with observed vs inferred values, builds Iconify icons into VectorDrawables, generates artwork instead of cropping the screenshots, implements MVI screens, then compares emulator or Roborazzi renders against the reference. Use whenever the user supplies screenshots, mockups or a design repository and wants them built as an Android Compose app, says "make this screen in Compose", "match the design", "px to dp", "screenshot to code", or wants a whole app UI built design-first, even if they never say "Compose".

Related Skills