Overview

Every rule here reads the file's header, not its pixels, so a few hundred bytes answer locally what a multi-megabyte upload would have answered. That is also more correct: a fully opaque RGBA PNG is still rejected by Apple, so sampling pixels for transparency would give the wrong answer.

Output is JPEG, deliberately: a canvas-exported PNG still carries a 32-bit alpha channel even when every pixel is opaque, so it would fail the very rule you came here to fix. JPEG cannot represent alpha at all, and both stores accept it. Resizing letterboxes rather than stretches — a distorted UI screenshot reads as broken to reviewers.

Apple indexes your app name, subtitle and the 100-character keyword field as one pool of words, then combines them. A phrase is reachable when each of its words is somewhere in the pool — the phrase itself never has to appear — a word already in your title is already indexed, and every space is overhead. The field is never shown to a user, so none of this is self-correcting.

Text overruns are the second most common cause of a blocked submission after malformed screenshots, and the limits differ per store — the same listing can pass Apple and fail Play. Fields above 90% of their ceiling are flagged too: translations commonly run 15–30% longer than English.

Only checks the store(s) above — uncheck one your app isn't on (Google Play for an iOS-only app, say) so a field that store requires but yours doesn't need stops being flagged. Remembered for next time.

Check multiple locales at once

Paste a {"locales": [...]} document — the same shape appmates validate-metadata reads on the command line — to check every localisation of a listing in one pass, instead of one at a time above.

Composed from whatever you've already entered on Screenshots, Keyword field and Listing text — nothing is re-typed here. Fill in any of those tabs and this one updates live, giving one combined go/no-go answer instead of three separate reports to read and reconcile by hand.

Apple and Google let you set a different price per storefront but give no guidance on what to set it to. This turns one price into a starting point per territory — a read-only sanity check to copy into App Store Connect or Play Console yourself. Nothing here ever calls either API.

Every ASO tool asks how hard is this keyword? as though there were one answer. There are 175 storefronts and the answer differs in each — difficulty tracks the language a term is searched in more than the size of the country. That makes it a launch decision: which locale to write first, which language to localise screenshots into.

Deliberately slow: one request per storefront against a free endpoint that rate-limits, spaced politely apart. Cached afterwards.

Who holds a term, what the field does with its screenshots, and the vocabulary everyone agrees on. Dimensions are read from the screenshot URLs themselves — the catalogue encodes the served size — so describing a whole gallery costs no bandwidth.

Every test you're running, App Store or Play, in one place — click one for its invite link and progress. Both stores gate that link the same way: a tester can't open it until you've added them by hand in that store's console first — Play Console's tester list, or App Store Connect's TestFlight testers. Nothing here can do that step for you; it only tracks where each test stands. Still need testers, or an app to test? That's a different page — Get testers / Be a tester is the actual two-sided marketplace, with real people requesting to join. This page is just your own manual log once a test is already underway, no account needed.