Your next build has a rejection in it.Find it before Apple does.
31 App Store checks your coding agent runs against your own repo — permissions, metadata, subscriptions, privacy, export compliance — with evidence for every result.
$99 one-time when the founding release opens. The list gets it first, at that price.
23 files of Markdown and JSON. No executable, no telemetry, no sign-in, no code changes.
What it finds
Read the full report →A Contacts permission is declared, but nothing in the project uses contacts.
app.json → NSContactsUsageDescription
Delete the key, or add the package that needs it. App Review flags unused permissions under Guideline 2.3.
1 deterministic blocker must be resolved before submission.
Checks the tool proved from your project and build.
External steps you confirmed. Your word, kept separate from what the tool verified.
Nobody has confirmed these yet. The report will not guess.
iOS app identity
iOS permission explanations
Expo package compatibility
App Store listing metadata
Subscription submission wiring
iOS build artifact (final Info.plist)
Developer-confirmed submission checklist
Internal dogfood build. Repository findings are real; checks requiring App Store Connect are explicitly marked needs-input when no export is supplied.
Download raw JSONGenerated from a deliberately imperfect Expo fixture. The report never claims your app will be approved — it separates what it verified from what still needs your confirmation.
How it works
Give the pack to your agent
Drop it in your repo and point Claude Code, Codex, Cursor or Hermes at it.
It audits, read-only
All 31 checks, each one passed, failed, or explicitly unresolved. Never a silent skip.
Fix, then submit
Every problem names its evidence, the guideline behind it, and the next action.
Scope, both directions
It does
- 31 checks: identity, permissions, compatibility, metadata, subscriptions, final build, export compliance, reviewer readiness.
- Every item resolved exactly once — pass, fail, developer-confirmed, or unresolved.
- Adapters for Claude Code, AGENTS.md, Cursor and Hermes. Otherwise it's plain Markdown.
It does not
- It never signs in to or reads your App Store Connect account. Repository wiring cannot prove what is attached to a version; the report says so instead of guessing.
- The package contains no executable, installer, account, or telemetry. Its protocol forbids network requests, source uploads, package installation, secret access, and repository changes during an audit.
- It is not affiliated with Apple, and it does not predict what a reviewer will decide.
Get founding access
$99 USD one-time$99USD
$99 one-time when the founding release opens. The list gets it first, at that price.
Founding build v0.2.0 — 31 evidence-backed checks. Claude Code ran the previous build on a real Expo app, made no network requests, and changed nothing in the project. Checkout opens when the last paid-launch checks pass.
Joining does not reserve a charge or create an order. Read the terms and refunds and the waitlist privacy terms.
What does the App Store preflight check?
It runs 31 evidence-backed checks across app identity, privacy permissions, purchases, account deletion, store metadata, reviewer readiness, final iOS artifacts, compatibility, and evidence integrity.
Does it work with Expo and React Native apps?
Yes. The founding pack targets Expo SDK 52, 53, and 54 plus React Native iOS projects. It includes native Claude Code discovery and adapters for other coding-agent workflows.
Does it upload my app or access App Store Connect?
No. The skill is a read-only audit protocol with no executable code, no network requirement, no credentials, and no App Store Connect access. External facts remain unresolved until you provide evidence.
Does a passing report guarantee Apple approval?
No. It catches avoidable submission blockers and separates verified facts from developer confirmations and unresolved checks. Apple alone decides whether an app is approved.
Can I use it with Claude Code?
Yes. Version 0.1.1 was installed and invoked through Claude Code's native skill discovery on both a synthetic Expo fixture and a real Verto Expo app snapshot. The current build is 0.2.0, which adds three checks plus a golden-fixture harness so results stay identical across runs and models.
How much will the founding release cost?
The planned founding price is $99 USD once, not a subscription. The waitlist is free and checkout remains closed until the remaining paid-launch quality and delivery gates pass.
When checkout opens, the versioned Markdown skill pack is delivered immediately after payment. If paid validation fails, Verto refunds every founding purchase in full.