Skip to content
Damjan Dabodabo.dev

Building in Public

I Open Sourced an App Store Rejection Checker for Claude Code and Codex

A free skill that audits your iOS app repo against the live App Store Review Guidelines and flags rejection risks before you submit. MIT licensed, on GitHub now.

Damjan Dabo
Damjan Dabo2 min read
iOSApp storeClaude codeAIIndie dev
A teal card titled App Store Rejection Checker, with a pre-submit check window listing App Store guideline rows marked pass or flag and a one issue to fix summary
Contents·3

I published the tool I built to catch App Store rejection risks before submitting. It's called app-store-rejection-checker: a skill for Claude Code and Codex that audits an iOS app repo against Apple's live App Store Review Guidelines. Free, open source, MIT licensed.

I've been shipping my own iOS apps since 2022: Itemlist, QRGenie, BarcodeCraft. Every one of those went through App Review, and the checks in this skill distill what those submissions taught me.

What it checks

The skill runs static checks over your codebase and project config: privacy usage strings, entitlements, privacy manifests and required-reason APIs, Info.plist, and your store metadata when you provide it.

The part I care about most: it fetches the live App Store Review Guidelines when it runs. So a finding cites the current guideline number, or the upload validation it would trip, instead of a stale copy of the rules baked into the tool. Apple updates the guidelines; a checker that memorized last year's version quietly drifts wrong.

The report ranks findings by severity: validator-blocking, review-risk, or noted. Each one comes with file and line evidence and the citation, so you can judge it instead of taking the tool's word for it.

Before publishing, I ran it on Itemlist's 2.1 release candidate. It caught real issues. Small ones, but real, in an app I've been shipping for four years.

What it can't do

It cannot guarantee approval. It audits what's in your repo, and review outcomes still depend on human reviewers. The README has an honest-limits section that says exactly this.

The skill itself was built and reviewed with Claude Code and Codex, which feels right for a tool that runs inside both.

How to use it

Give Claude Code or Codex the link to the repo and ask it to check your app. It handles the install itself:

Install the skill from https://github.com/dabodamjan/app-store-rejection-checker and run it on this repo.

If you're about to submit an iOS app, run it first. If it finds nothing, you upload with a bit more confidence.

Written by

Damjan Dabo

Writing field notes from Croatia 🇭🇷

More about Damjan

Continue reading

Building in Public

Claude Code Made My App Store Screenshots in 13 Languages

Generating fully localized App Store screenshots with Claude Code and Codex: the complete Itemlist set across 14 storefronts, review gates that caught real bugs, and my only job was feedback.

Building in Public

A Much Cheaper Way to Run Fable, the Most Expensive AI Model

Fable is the best AI model I've used and the most expensive, double Opus. When the free window ends, you pay by the token. Here is how I cut its cost per turn by nearly a third without giving up the model.

Building in Public

48 Hours With the Best AI Model in the World: My Experience

Anthropic put Fable, their priciest model, free on Claude Max. I let it loose on my own sites and asked Claude Code what it cost. About a thousand dollars.

Newsletter

You'll know when I ship a new post.

Just the post. No marketing drip, no upsell ladder. Unsubscribe in one click.