Storewatch gives the AI you already use the numbers, listings, reviews, keywords and prices of every app you ship. Ask it anything, or ask it to change something. Every change is a draft until you publish.
Ask the way you'd ask a teammate who knows every number. Here are a few things people ask.
Titles, subtitles, keywords, descriptions and what's new, in every language, plus the keywords you track.
Crashes after a release, listing problems, prices that drifted and countries you're not on sale in.
Downloads, conversion, proceeds, subscriptions and reviews for one app or all of them, over any period.
The AI reads your data through read-only tools. Anything it wants to change comes back as a draft that shows exactly what will change.
| What it can change | When it reaches Apple |
|---|---|
| Listing text in every language: title, subtitle, keywords, description, what's new | When you publish, from the chat or the Listing screen |
| Prices of your app, in-app purchases and subscriptions, per country | When you apply them on the Pricing screen |
| Replies to reviews, in the reviewer's language | When you click Send |
| Keywords you track, events on your charts, insights you've dealt with | Never: these stay in Storewatch on your Mac |
Use one for everything, or a different one for each job: chat, review replies, translations, listing checks and written reports. You can even compare two side by side.
Anthropic's models with your own API key. Pay per use.
OpenAI's models with your own API key, separate from a ChatGPT Plus subscription.
Google's models with a key from Google AI Studio.
Uses the claude command on your Mac and your Claude subscription. No key needed.
Uses the codex command on your Mac, signed in with your ChatGPT account.
One key for many models from different companies.
Open models running on your Mac. Nothing leaves it.
Free, private and offline. Needs macOS 26 with Apple Intelligence.
Everything else in Storewatch works without it.
Storewatch has no servers that see your App Store data. What the AI sees depends on the AI you pick.
Claude, ChatGPT, Gemini and OpenRouter get your question and only the data needed to answer it, straight from your Mac, under your agreement with them.
Claude Code and Codex run on your Mac and talk to their provider under your own account with them.
Ollama and Apple's on-device model run locally. Nothing is sent anywhere.
API keys are kept in your Mac's Keychain. Details are in the Privacy Policy.
Ask the Claude app, Claude Code, Codex or Cursor about your apps without opening Storewatch. Connecting takes one click, or one pasted prompt.
Storewatch doesn't charge for AI. You pay your provider for what you use with your own API key, use a subscription you already have through Claude Code or Codex, or run a model on your Mac for free with Ollama or Apple's on-device model.
No. It reads your data through read-only tools. Listing text, prices and review replies arrive as drafts that show exactly what will change, and nothing reaches the App Store until you publish, apply or send it.
The one you already use. Claude, ChatGPT, Gemini, OpenRouter and Ollama look things up in your data as they go. Claude Code and Codex use the subscription you're signed in with and get a summary of your data with each question. Apple's on-device model is free and private and needs macOS 26 with Apple Intelligence. You can pick a different AI for each job.
No. Insights, analytics, reviews, keyword ranks, releases and prices all work without it. AI adds chat, drafts and written reports.
