Apple App Store Reviews API for AI Agents: Pull iOS Ratings via MCP
How do you pull thousands of Apple App Store reviews for an app without babysitting a browser or copy-pasting star ratings by hand? The Apple App Store Reviews API answers that: give it an app and it returns iOS and macOS reviews as clean, structured JSON, ready for analysis, sentiment work, or an AI agent to reason over. If “app store reviews api” is the thing you have been searching for, this is the builder-friendly, MCP-first version of it.
What the App Store Reviews API does
It fetches Apple App Store reviews on demand and hands them back as structured JSON. Each review carries the fields you actually want to work with: the star rating, the review title and body text, the review date, the app version it was left on, the author, and the helpfulness count. It reaches across 50+ country stores, so you can compare how the same app is received in the US, Japan, the UK, and beyond. You can sort by most recent, most helpful, most favorable, or most critical, which makes it easy to pull exactly the slice you care about. Pricing is pay-per-result, starting from $0.01 per 1,000 results, so a quick pull of a few hundred reviews costs a rounding error.
Get JohnVC’s stories in your inbox
Join Medium for free to get updates from this writer.
One honest limitation worth knowing up front: the API reads what Apple publishes on the public store, so coverage and freshness track the store itself, and very old reviews on high-volume apps are paginated (an unlimited pull is capped internally at 50 pages). You pass either a numeric App Store product ID or just the app name and let the actor resolve it.
Who it is for
ASO and mobile marketers who need review volume, sentiment, and version-tagged feedback to move the needle on store ranking. Product managers and indie founders tracking what users love and hate after each release. Data scientists and analysts running sentiment or topic models over thousands of reviews at once. And AI agent builders who want their assistant to fetch and summarize real customer feedback as a tool call, no scraping code required.
Ready-to-run examples
The actor ships with four example tasks, each one runnable in a single click and easy to clone and tweak for your own app. You can browse them all on the App Store Reviews API examples page, or jump straight to the ones below.
Export TikTok App Store reviews to CSV
Pull TikTok’s App Store reviews and drop them into a spreadsheet-ready CSV in one run. This is the fastest way to get a big, well-known app’s feedback into a format your analysts or a BI tool can chew on immediately. Great as a template: swap the app and you have any title’s reviews on tap. Try the export TikTok reviews to CSV example.
Pull a competitor app’s one-star reviews
Sort by most critical and pull only the one-star reviews for a competitor. That is where the sharpest product signal lives: the exact complaints, missing features, and dealbreakers that are pushing their users away and might win them over to you. Run the competitor one-star reviews example.
Analyze your iOS app reviews in Claude via MCP
Wire the actor into Claude as an MCP tool and just ask, in plain English, for your app’s latest reviews and a summary of the themes. Claude discovers the tool, runs it, and reasons over the JSON it gets back, so you go from question to insight without writing a scraper or a parser. See the analyze reviews in Claude via MCP example.
Export App Store reviews to CSV
The general-purpose version of the export task: point it at any app by product ID or name and get a clean CSV of reviews, ready for Excel, Sheets, or a pandas dataframe. This is the one most people start from when they just need the data out and into their own tooling. Open the export App Store reviews to CSV example.
Getting started
The input is small. You give it an app (a numeric product ID or an app name), a country store, a sort mode, and how many reviews you want. Here is a typical run that pulls the 100 most critical reviews for an app:
{
"product_ids": ["835599320"],
"country": "us",
"sort": "mostcritical",
"max_reviews": 100
}Want a multi-app competitor sweep in one run? Pass a list of product IDs:
{
"product_ids": ["324684580", "284882215", "447188370"],
"country": "us",
"sort": "mostrecent",
"max_reviews": 200
}Set max_reviews to 0 to drain everything available, and flip include_macos on for Mac apps. You can run it from the Apify Console, over the Apify API, or from any MCP client. To call it from Python with the Apify client:
from apify_client import ApifyClient
client = ApifyClient("YOUR_APIFY_TOKEN")
run = client.actor("johnvc/apple-app-store-reviews-api").call(run_input={"app_name": "spotify", "country": "us", "sort": "mostrecent", "max_reviews": 200})
for review in client.dataset(run["defaultDatasetId"]).iterate_items():
print(review["rating"], review["review_title"])To use it as a tool in Claude, Cursor, or another AI client, point the client at the hosted Apify MCP endpoint:
https://mcp.apify.comThen ask your agent something like “pull the 100 most-recent App Store reviews for Spotify in the UK store and bucket them by star rating,” and it will discover the actor, run it, and hand you back the structured result. Export formats include JSON, CSV, and Excel from the dataset.
FAQ about scraping App Store reviews
Should I use the App Store Reviews API or a web scraper?
If you want reviews as clean JSON without maintaining anything, the API is the easier path. A hand-built scraper means writing parsers, handling pagination across country stores, and fixing it every time the store markup shifts. The App Store Reviews API gives you the same result as a scraper, ratings, titles, text, dates, and versions, but you just pass an app and get structured data back.
Does the App Store reviews scraper work with MCP and AI agents?
Yes. It is built MCP-first, so you can register it as a tool in Claude, Cursor, or Claude Cowork through the hosted Apify MCP endpoint and let the agent call it. Your assistant can fetch reviews for an app, then summarize sentiment or extract feature requests in the same conversation, without you writing any scraping code.
How is this different from building my own App Store review scraper?
A DIY scraper is code you own forever: request handling, HTML parsing, retries, proxy rotation, and the ongoing maintenance when Apple changes its pages. This actor is a maintained, pay-per-result service starting at $0.01 per 1,000 results, so you skip the build and the upkeep and get versioned, structured reviews across 50+ stores from day one.
Can I schedule this App Store reviews scraper run?
Yes. From the Apify Console you can put any task on a daily or weekly schedule, so a fresh batch of reviews lands automatically without anyone clicking run. That is ideal for tracking a release over time, watching a competitor, or feeding a sentiment dashboard. Set it up on the App Store Reviews API and let it run on its own.
Start pulling reviews
If you have been looking for a dependable app store reviews api that returns structured iOS and macOS feedback and plugs straight into your agents, this is it. Grab an app ID or name and run the Apple App Store Reviews API on Apify, or clone one of the ready-made examples to see it work in a single click.
