agentleFS
mikezamayiasmikezamayias/agent-skills/skills/store-launch-mastermind/SKILL.md

store-launch-mastermind

Skill1 starsChanged 27 days agoBSD-3-ClauseView on GitHub

Orchestrates end-to-end public go-live readiness for App Store and Google Play apps, ending in a go or no-go decision. Use when launching a Flutter, Swift or Kotlin app or running store health checks, including Mac direct release. Also covers assets, metadata, privacy, legal docs, ratings, Data safety, billing, TestFlight and rollouts.

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 7 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this as the top-level launch orchestrator.

    line 11
  2. Distinctive:

    Ask or infer, then confirm:

    line 19
  3. Distinctive:

    Check the current Apple pages before planning a launch around them.

    line 30
  4. Very specific:

    Run launch-readiness for code, policy, store-compliance, security, accessibility, localization, performance, and polish gates.

    line 39
  5. Distinctive:

    If a guideline interpretation is borderline, the user can book a Meet with Apple appointment before submitting.

    line 42
  6. Distinctive:

    Only the build, signing and crash tooling change.

    line 56
  7. Distinctive:

    Check their bundle IDs and entitlements as part of the iPhone build.

    line 68
  8. Distinctive:

    Stop and ask for explicit user confirmation before:

    line 83
  9. Very specific:

    Write or update tasks/store-launch-report.md in the project root when working inside a repo.

    line 93

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Store Launch Mastermind

Use this as the top-level launch orchestrator. Do not inline every store workflow; route specialized work to the focused skills and collect their results into one go/no-go report.

References

  • Load references/phase-map.md before planning or running a launch.

Required Intake

Ask or infer, then confirm:

  • Project type per target: Flutter, native Swift (iPhone, Watch, widgets, Mac) or native Kotlin (Android).
  • Target stores: App Store, Google Play, or both.
  • Mac distribution: Mac App Store, or Developer ID with notarization and a direct download or Homebrew cask.
  • Current phase: first app creation, beta/internal testing, metadata/assets, review submission, approved-but-not-published, public rollout, or rejection fix.
  • App identifiers: bundle ID, package name, SKU, app ID, version/build/version code.
  • Monetization: free, paid, IAP, subscriptions, ads, external physical goods/services.
  • Sensitive features: kids/family, UGC, social, health, location, camera/photos, finance, government, gambling, crypto, VPN, news, AI-generated content.
  • Features that depend on Apple Intelligence or Siri AI: record the regions, languages and OS betas they need, because Siri AI is not available in the EU on iOS and iPadOS, lists six languages, and the Greek app locale is not one of them (Apple pages, Nov 3, 2026 post). Whether a developer located in the EU can test it on an iPhone is a single MacRumors report and not an Apple statement. Check the current Apple pages before planning a launch around them.
  • Data practices: accounts, analytics, crash reporting, tracking, third-party SDKs, data deletion, privacy policy, ToS.
  • Asset state: icons, feature graphic, screenshots, app preview/video, App Store creative assets (product page header, search results), localized listing text.
  • Deadline and risk tolerance.

Orchestration Flow

  1. Discover project stack and store state. Detect Flutter (pubspec.yaml), native Apple (project.yml, *.xcodeproj, Package.swift) and native Android (settings.gradle.kts, gradlew) as launch-readiness does, and run every later step per project.
  2. Run launch-readiness for code, policy, store-compliance, security, accessibility, localization, performance, and polish gates.
  3. Route asset gaps to app-graphics and app-screenshots.
  4. Route App Store operations (release flow, submission health, metadata sync, ASO audit, screenshot upload) to the installed App Store Connect tooling, such as the asc CLI or skills built on it. If a guideline interpretation is borderline, the user can book a Meet with Apple appointment before submitting. Where: the Meet with Apple schedule page on developer.apple.com. Steps: request a slot and include the app's Apple ID, if one exists, and the specific questions. Cost: a 30-minute Webex video call on Tuesdays and Thursdays, and slots can fill quickly (two X posts report full schedules, both unverified). Open the schedule page a few weeks before the planned submission, because Apple's page states no date for new slots. The appointment is informational only and does not pre-approve or expedite a review.
  5. Route Google Play operations to the installed Google Play Console tooling, such as the Play Developer API or Fastlane supply.
  6. Route paywall/subscription concerns to paywall-conversion-flow, subscription localization in App Store Connect, RevenueCat/catalog skills when applicable, and Play Billing checks.
  7. For privacy policies, terms, consent copy, appeal letters, or legal letters, involve legal review skills if available and clearly label the output as drafting support, not legal advice.
  8. Produce a single launch report with verdict, blockers, manual gates, exact commands, and owner/action/date.

Native Swift and Kotlin projects

The flow is the same. Only the build, signing and crash tooling change.

StepFlutterNative AppleNative Android
Generate projectflutter pub getxcodegen generate when project.yml existsnone
Testsflutter testxcodebuild test -scheme <TestScheme> -destination '<dest>', swift test in packages./gradlew :app:testDebugUnitTest
Release buildflutter build ipa, appbundlexcodebuild archive -scheme <App> -destination 'generic/platform=iOS' -archivePath <path>./gradlew :app:bundleRelease
Uploadasc or Fastlaneasc builds upload or the asc-xcode-build skillPlay Console or Play Developer API
Crash sourceSentry or CrashlyticsMetricKit, App Store Connect crash reports, asc-crash-triage, or the project's SDKAndroid vitals in Play Console, or the project's SDK
SubscriptionsRevenueCat or in_app_purchaseStoreKit 2 or RevenueCatGoogle Play Billing or RevenueCat
  • A watchOS app and its widgets ship inside the iPhone archive and are reviewed with it. Check their bundle IDs and entitlements as part of the iPhone build.
  • A Mac app outside the App Store has no review. Its gate is Developer ID signing, hardened runtime, notarization, a stapled ticket and the cask or download page, per launch-readiness and asc-notarization. Skip the Mac App Store checklist for it.
  • Android needs JDK setup from the project's own docs before ./gradlew.
  • Without a crash SDK, "crash reporting" in the intake means store-provided reports, so confirm someone reads them.

Verdict Rules

  • BLOCK: legal/policy blocker, missing privacy policy/Data safety/App Privacy, build cannot be submitted, missing required assets, payment policy violation, target SDK/API blocker, app crashes, or required account access/test credentials missing.
  • HOLD: no hard blocker, but high-risk manual declarations, weak assets, missing localization, ASO gaps, unverified subscription restore/purchase flow, or incomplete staged rollout plan.
  • SHIP: all hard gates clear, manual store forms confirmed, assets and metadata complete, build/release status verified, and rollback/pause plan documented.

Manual Gates

Stop and ask for explicit user confirmation before:

  • Submitting for review.
  • Publishing or starting production rollout.
  • Changing price, availability, subscriptions, base plans, offers, or IAPs.
  • Declaring Data safety, App Privacy, age/content ratings, target audience, ads, health, finance, or other legal/policy claims.
  • Sending appeal/legal letters.

Output Format

Write or update tasks/store-launch-report.md in the project root when working inside a repo.

Include:

  • Verdict: SHIP, HOLD, or BLOCK.
  • Store matrix: App Store status, Google Play status.
  • Blockers and high-risk items.
  • Required manual declarations.
  • Assets checklist.
  • Metadata/localization checklist.
  • Build/release commands already run and their result.
  • Next commands or UI steps.
  • Rollback/pause plan.

Sources

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.