App Store Screenshot Playbook

Turn listing views into installs: a seven-frame screenshot plan built from what the top apps in your category ship.

The job: keywords get you seen. This gets you installed.

Someone lands on your listing and makes up their mind in under seven seconds. In that time they look at three screenshots and maybe two seconds of a muted video. Most listings burn the window on a "Welcome to AppName" opening shot and captions nobody reads. This skill works it the other way round: pull what the winning apps in your category actually ship, find the look they all share and the one that breaks it, then build your seven frames against that.

What's on the tag:

The competitor picture atlas. Apple publishes the screenshot image links for every app on the store, and a documented trick upgrades those links to full size. Your agent downloads the top-ranked sets, looks at them, and answers seven questions about each one — the opening shot, the caption style, the framing, the colours, where the best moment lands — until you can see the category's normal, and its one outlier.
The seven frames, in order: opening shot, what makes you different, the moment the app earns its keep, two everyday uses, a peek at Pro, then the close. Plus the swaps for apps with no paywall, and the rule about which single frame is allowed to show a price.
Captions that survive the scroll. Four to eight words, verb first, the result rather than the feature — with a table of patterns and a list of the dead phrases. "Powerful features" is noise. "AI fills in the details" is a caption.
The preview-video plan: a four-part script structure, hook ideas sorted by category, and a template you fill in.
How to test a new set before you commit to it, using Apple's Custom Product Pages. One change at a time, fourteen days, enough views to mean something. Includes the documented case where an indie founder's own homemade screenshots beat the professionally designed set — which is why this skill makes you test first instead of trusting the polish.
Eight other countries, done properly. Three levels of effort, plus the money and date format tables that keep a German or Japanese screenshot from quietly costing you trust.
Shooting the screens off the simulator: how to grab them at the device's full size, how to set the clock to 9:41 the way Apple's own shots do, and how to load the app so it shows screens a real user would need days to reach.

Who it's issued to: developers whose numbers show people finding the listing and walking away, and anyone building a screenshot set without a designer on payroll.

Why not a free directory download? Free skill files tell you screenshots matter. This one was run against live listings — why the frames have to read at thumbnail size, the full-size image trick, the line between learning from a competitor's set and copying it — and it gets updated as Apple moves the limits. Buy once, keep the latest version in your locker.

FIELD REPORT real output, not a promise

Two artifacts from the playbook's own documented material: the consensus/breakout read the atlas step produces for a habit-tracker category, and a per-slot recommendation in the exact handoff format the skill outputs.

Atlas output — habit-tracker category (Step D)

After downloading 7 competitors' screenshot sets via iTunes Lookup and running the 7-question analysis:

Consensus: floating phone with brand-color gradient background, 5-word bold-only caption, slot 1 = hero with the full app shown, slot 3 = streaks/calendar visual as the magic moment. 6/7 use the 9:41 status bar. 0/7 show pricing. 0/7 use human photos.

Breakout (the top-ranked indie): no device frame, full-bleed app screen, caption ABOVE the screenshot in a soft pastel, slot 1 = a near-empty state with a single completed habit rather than a maximalist dashboard. This is what "honest indie" reads as in the category — and it's the top converter.

Per-slot recommendation (Step E format)

Slot 1 (Hero):
  Consensus: floating phone + brand gradient
  Breakout: full-bleed app screen, no frame
  Recommendation: adopt the breakout. Reason: this app's UI is already distinctive,
  so showing it bare IS the differentiator. Caption: "Build a habit that sticks" —
  outcome-led, 5 words.

The same structure repeats for all seven slots, so the plan hands straight to a designer (or a rendering pipeline) without interpretation.

The A/B lesson baked into Step 5

From the documented indie habit-tracker case the skill cites: the founder A/B-tested professionally designed screenshots against his original self-made set via Custom Product Pages — and the self-made set won on conversion. The playbook therefore hard-codes the rule: never swap to polished creatives without running the CPP test first. The polish bias is real, and the data gets the veto.

SERVICE RECORD living gear — updated as the factory learns

v1.1.0 — 2026-08-25

Field-tested capture and footage upgrades, learned shipping store sets and app ads across a live iOS portfolio this summer. New: a data-staging discipline for quantity-driven apps (seed the device, let the app compute every number, then use it — seeding doubles as a load test that has caught a ship blocker), the headless simulator recording recipe with eight silent traps that all exit 0 (VFR seeking lies, dropped beats, tpad discarded next to -to, -v error muting the measurement filters, a zsh modifier eating font paths, and more), and a new reference on social-ad footage — where the App Store preview rules invert: the four-beat one-feature ad, VO-first build order, payoff-first structure, honesty rules for voice-over claims, and 9:16 conversion. Also new in Step 8: pick simulators by pixel size not name, reboot before capture (the stale ◀ OtherApp back-link), entitlement-state staging via launch arguments, watchOS companion screenshots, and the regenerate-and-diff rule for script-rendered sets.

v1.0.0 — 2026-07-17

First issue. Ported from the factory's internal skill: sanitized for general use, methodology intact, field report captured from a real run.

Every update ships free to owners — your locker always serves the latest version.

QUESTIONS

Do I need design tools or a designer?›

No. You get the plan — which screen goes in which frame, the caption for each, and the reason behind both — plus the commands to capture the real screens. Build the finished frames wherever you like: Figma, fastlane, any framing tool. The size specs and Apple's rules come with it.

Does it really pull competitors' actual screenshots?›

Yes. Apple publishes the screenshot image links for every app on the store. The skill swaps the size in each link for the full-size one, downloads each competitor's set in order, and works through the same seven questions on all of them.

Is testing two screenshot sets against Apple's rules?›

No. It uses Custom Product Pages, which is Apple's own feature for running more than one version of a listing. You get the setup steps and the rules for reading the result honestly instead of talking yourself into it.

What agents does it run on?›

Written for Claude Code, which can also look at the downloaded images itself. It also works in claude.ai Projects, and in Codex setups that read an AGENTS.md file. Every file is plain markdown.

one-time · v1.1.0 · updated 2026-08-25

Instant download (.zip) · works with Claude Code, claude.ai projects & Codex · 7-day refund, no forms.

WHAT'S IN THE CRATE

▸See exactly what the top apps in your category are showing. Your agent pulls their real screenshots from Apple's own public listing data, upgrades the image links to full size, and answers seven questions about each set — so you end up with the look everyone copies, and the one app that breaks the pattern
▸Know what belongs in each of the seven frames: the opening shot, what makes you different, the moment the app earns its keep, two everyday uses, a peek at Pro, and the closing ask. Swaps included for apps with no paywall and apps that lead with reviews
▸Write captions people actually read. Four to eight words, verb first, the result rather than the feature — with a table of patterns for each frame and the list of dead phrases to cut
▸Script the preview video: a four-part structure for 15 to 30 seconds, hook ideas sorted by category, what to do about sound, and a fill-in-the-blanks template
▸Record the footage with no camera and no crew. The exact commands to record the simulator and trim the clip, plus eight ways the job goes wrong while every command still says it worked. Then the four-beat single-feature ad: voice-over first, payoff up front, honest claims, and the tall 9:16 crop for TikTok and Reels
▸Test a new screenshot set before you commit to it, using Custom Product Pages — Apple's own way of running more than one version of a listing. Setup steps, what to test and in what order, and the discipline that keeps the answer real: one change at a time, fourteen days, roughly 5,000 views of each version
▸Ship the same set in eight other countries without looking sloppy. Three levels of effort, captions rebuilt in the language instead of translated word for word, and the money and date format tables that stop the obvious mistakes
▸Stay inside Apple's rules for store artwork: the 9:41 clock in the status bar, no prices on feature shots, no made-up content, the size specs, and the icon rules
▸Fill the app with the right data before you shoot. Seed the device so every number on screen looks real and earned, pick a simulator by its pixel size rather than its name, shoot the same screen twice — once as a free user, once as a paid one — using a flag you hand the app when it starts, grab the Apple Watch companion shots, and edit saved data directly for screens a real user would need days to reach

ISSUED TO

▸People are finding your listing and not installing, and you don't know why
▸You're building a screenshot set on your own, with no designer
▸You're launching in other countries and the pictures need to keep up with the keywords

Also issued in:

The App Store Kit — $29 The iOS Factory — $69