- Home
- Skills
- Mobile Development
- App Store Release Kit: iOS Scans, IAP Setup, Review Prompt
More screenshots
Works with the AI Agents you already use
App Store Release Kit: iOS Scans, IAP Setup, Review Prompt
**Setup is nothing more than Python 3.** The scans and the icon maker are single Python files, and the in-app purchase script uses only Node's built-ins.
$19
App Store Release Kit: iOS Scans, IAP Setup, Review Prompt
Example session with this skill installed
I'm about to send my Expo app to TestFlight for the first time. Check it for anything that would get it rejected, and set up my two in-app purchases (Remove Ads $1.99 and 500 Coins $0.99).
- Read your context and instructions
- Compiled the app store release
The agent runs python3 scripts/scan_scaffold.py . in your app folder. It reports, for example:
- 3 errors:
EXPO_PUBLIC_STRIPE_SECRET_KEYin.env, which would be built into the app where anyone can read it- the Expo default bundle ID
com.anonymous.demoappinapp.json NSAllowsArbitraryLoadsswitching off App Transport Security
- 3 warnings: the leftover
HelloWavestarter component (on two lines ofapp/index.tsx, one warning each) and alocalhost:3000API address in app code
Using fixing-findings.md, it proposes each fix: move the secret to your server, register your own bundle ID (never changing it once live), and read the server address from build configuration.
For purchases, it writes iaps.json from the included example and runs a dry run that prints the plan: two products, their type, price and locales, and "Nothing was sent". It runs --apply only after you confirm, with your API key in environment variables.
After your build, it runs the privacy manifest check on the .app and lists any UNDECLARED required-reason categories to add to PrivacyInfo.xcprivacy. It finishes with the pre-submission review prompt: a pass/fail verdict and a table of problems with file and fix. You press Submit yourself.
Connects securely to your tools. The creator never sees your data.
About this skill
Catch the avoidable rejections before App Review does. Your agent runs one command on your project and gets a list of the things that get first submissions bounced or leave an app exposed:
- a framework-default
com.exampleorcom.anonymousapp ID - starter screens nobody replaced
- a server key shipped inside the app
- App Transport Security switched off
- a missing privacy-manifest reason
- an icon with an alpha channel
For each finding, references/fixing-findings.md tells the agent how to fix it safely and which changes break a live app.
Setup is nothing more than Python 3. The scans and the icon maker are single Python files, and the in-app purchase script uses only Node's built-ins. No npm packages, and Pillow only for the icon. Run them from your app folder.
What you get
- Project scan (
scan_scaffold.py, 14 rules at error / warn / info level):- framework-default identifiers and names
- Expo, Flutter and React Native starter leftovers
- server-only database keys
- secrets in
EXPO_PUBLIC_variables - private keys (reported without echoing key material)
- disabled ATS and Android cleartext
- open Firebase rules
- dev hosts in app code
- OTA-update and web-wrapper markers
- Privacy manifest check (
scan_privacy_manifest.py) on the built.apporPayload. It finds required-reason APIs referenced by each Mach-O binary and checks them against the nearestPrivacyInfo.xcprivacy, so each framework is judged by its own manifest. It also flags reason codes from the wrong category. The reason list was checked against Apple's documentation on 4 November 2026. - App icon maker (
make_app_icon.py). One square image of at least 1024 px goes in. An opaque 1024 icon and a single-sizeAppIcon.appiconsetcome out. It refuses to upscale. - In-app purchase creator (
create_iaps.mjs). Products, localizations and prices come from one JSON file. It is a dry run by default and safe to re-run, it never changes an existing price, and it signs its own ES256 token with your App Store Connect API key from environment variables. - Prompts. The pre-submission review prompt is organised by App Review Guideline number and cites a number only after checking the current guidelines. The featuring-readiness prompt is for after launch.
- CI templates. A fastlane
Fastfile(TestFlightbetalane, plus areleaselane that uploads withsubmit_for_review: false) and a Codemagic workflow for Expo.release_preflight.shstops a CI build on missing or placeholder signing settings without printing secrets.
Nothing submits your app. The only network call is create_iaps.mjs --apply, to api.appstoreconnect.apple.com, with your own key. 52 tests are included (29 Python, 7 Node, 16 preflight), and no test touches the network.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Mobile Development.
- The scans find known patterns. A clean scan is not a guarantee of approval: App Review decisions such as 4.2 (Minimum Functionality) and 4.3 (Spam) are judgment calls no tool can make.
- The privacy check matches API names in each binary's string table. A match shows an API is referenced, not that it is called; confirm before declaring a reason.
- Apple updates the required-reason list and the review guidelines. Both were checked on 4 November 2026.
create_iaps.mjshandles consumable, non-consumable and non-renewing subscription products. Auto-renewable subscriptions use a different API and are not covered.- The scanner reports each occurrence, so a name that appears twice on one line is listed twice.
- Using the fastlane or Codemagic templates means copying
scripts/into your app folder.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 12 days ago
- Software
Python 3.9+Bashexpo
- Optional
- Pillow
Node.js 18+XcodefastlaneCodemagic
Pillow is only for the icon script and Node.js only for the in-app purchase script.
Permissions
The scans never change the project and never send anything. Credentials are read from the environment, never written to files or printed. Nothing in the kit submits an app for review.