How to use the Apple App Store Reviewer agent skill

What does the Apple App Store Reviewer skill make the agent do?

Once installed, the agent works as an App Store Review Specialist that audits an iOS app's source code and metadata from the point of view of an App Store reviewer. Its goal is to identify likely rejection risks and optimization opportunities, not to predict Apple's decision.

The skill tells the agent to check files such as Info.plist, entitlements, privacy manifests (PrivacyInfo.xcprivacy), StoreKit configuration, onboarding flows and paywalls. It also covers account features such as login, Sign in with Apple and account deletion, and content features such as user-generated content, reporting and blocking. The output is ordered by priority, with the highest rejection risks first.

• Flags missing or incorrect permission usage descriptions and privacy disclosure gaps.
• Checks in-app purchase flows, including restore purchases and subscription handling.
• Looks for login walls, demo or test accounts for reviewers, and reviewer notes.
• Reviews user-generated content for filtering, reporting, blocking and content removal.
• Checks pricing wording in app metadata and screenshots.

How do you install the skill?

Download the ZIP file for the skill from this page. The package contains a single file, SKILL.md. You can install it in one of two ways.

For Claude Code, unzip the download into your personal skills folder at ~/.claude/skills/ so it is available in all your projects, or into a project's .claude/skills/ folder so it applies only to that project. In the Claude apps, you can instead upload the skill under Settings → Capabilities → Skills.

• Personal install: put the unzipped skill folder in ~/.claude/skills/.
• Project install: put it in the .claude/skills/ folder inside your iOS project.
• Claude apps: upload the ZIP under Settings → Capabilities → Skills.

How do you trigger the skill with a natural request?

You do not need a special command. Ask the agent in plain language to review your app for App Store rejection risks, and make sure the agent can see your iOS project. The skill is designed to read the project files, so point the agent at the repository it should inspect.

Naming the kind of check you want helps the agent focus. For example, you can ask it to check privacy, in-app purchases, or account deletion specifically.

• "Review this iOS app for likely App Store rejection risks before I submit."
• "Check Info.plist, entitlements and PrivacyInfo.xcprivacy for App Store privacy issues."
• "Audit our StoreKit and paywall code for restore purchases and pricing problems."
• "We got a Guideline 1.2 rejection. Check our reporting, blocking and moderation features."
• "Check whether account deletion is available and whether our login flow meets Sign in with Apple rules."

What should you expect from a review?

The skill tells the agent not to change any code during this first pass. It should read files, report findings and suggest improvements, and leave edits to you. It also asks the agent to separate verified findings, which it can point to in code or configuration, from assumptions and checks it could not confirm.

Apple's App Store Review Guidelines change over time. If the agent has internet access, the skill asks it to check Apple's current wording before quoting a guideline or stating that a requirement is current. Without internet access, treat the guideline references as possibly outdated.

• Expect a prioritized list of risks and recommendations, not a pass or fail verdict.
• Expect assumptions to be labelled as such.
• Expect references to guideline topics; check the current wording on Apple's site yourself before acting on them.

What checks are included for user-generated content and pricing?

The skill includes detailed checks that apply only when they are relevant to your app. For user-generated content under Guideline 1.2, the checks include terms that users accept before they can register or sign in, a zero-tolerance statement, unchecked consent controls, handling of reports within 24 hours, removal of content and ejection of the user who posted it, developer notification when a user is blocked, and instant removal of blocked users' content from the feed.

For pricing metadata under Guideline 2.3.7, the skill treats words such as "Free" or a sale price in the app name, subtitle, keywords or screenshots as pricing information to remove. It also asks the agent to check text rendered inside screenshot UI, not only captions. Screenshot checks are marked unverified if the screenshots are not in the repository.