Real-time product, search, and offer data across all Amazon marketplaces — without owning the scraping problem.
Type any ASIN. We'll fetch live data from Amazon and return structured JSON. This is exactly what your code receives.
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Every field is parsed, named, and ready to use
No HTML to scrape, no fields to extract. The same response shape across all marketplaces.
60+ clean fields per product. Already parsed.
Title, price, BSR, every seller on the listing, reviews, UPC/EAN, images, dimensions, deal and promotion state — returned as predictable JSON. No DIY HTML parsing, no breakage when Amazon changes a div.
Not a startup's first-attempt scraper.
The same Amazon-data infrastructure that's been running ASINSpotlight's desktop app for years, with thousands of paying users. Layout changes, marketplace-specific quirks, reliability at scale — all already solved. Years of keeping this data flowing reliably are the moat that doesn't show up in a feature comparison until your scraper breaks at 2 a.m.
Same call. Just change marketplace=.
US, UK, JP, DE, FR, IT, ES, NL, CA, MX, BR, AU, IN, SA, AE, SG, IE, BE — every Amazon domain, identical response shape. Most competitor APIs are US-first and charge per region. If your product crosses borders, ours follows.
Tell ChatGPT or Claude what to build. Working code in one session.
REST + API key. Predictable JSON. Self-describing field names. Complete OpenAPI spec the LLM can read. Working code examples in Python, Node.js, and PHP. The friction between "I know what I want" and "it's running in production" collapses to one prompt.
Your repricer / inventory tool / dashboard depends on prices, BSR, sellers, identifiers — and right now you're the one keeping the scraper alive. Every Amazon layout change is a fire drill.
A list of ASINs you re-check daily for price and Buy Box changes; keyword scans you run on a cadence; per-ASIN data you want flowing into a sheet or a database without you opening the app each morning.
You build with ChatGPT, Claude, or Cursor. You need the AI to wire up Amazon as a data source on the first try.
Five endpoints. Multiple use cases. Combine them however you need.
60+ fields per ASIN. Sub-second response. Marketplace param picks the domain. Deal state included: badge type, Amazon's deal ID, exact end time, and savings against the list price. Use it as the enrichment step after discovery, or to refresh data on products you're already tracking.
Builds: product cards in seller tools, daily price snapshots, identifier lookups for cross-marketplace listing.
Submit keyword + marketplace, get paginated results. Each entry already includes price, rating, reviews, monthly demand, deal badges, ASIN — enough to shortlist without a follow-up call.
Builds: product discovery flows, keyword research dashboards, niche scanners.
The full Buy Box and offer panel: every seller with price, condition, fulfillment method, rating, and delivery promise. Updates in real time, not from a cached database.
Builds: Buy Box monitors, seller-tracking tools, repricing decision logic.
Look up a seller storefront by its ID: name, 12-month feedback rating and positive %, total ratings, plus the legal business name and address from Detailed Seller Information when Amazon publishes it.
Builds: supplier vetting, brand-protection sweeps, reseller and counterfeit investigations, KYC on third-party sellers.
For deal pages, search-result variants, marketplace-specific layouts that don't fit a clean endpoint. Submit a URL, get structured JSON back. Same reliable infrastructure as the typed endpoints — just pointed at a URL you supply.
Builds: edge cases without writing a parser per page type, ad-hoc scrapes, scripts that adapt to page formats your code shouldn't have to know about.
Full customer review text by ASIN, as structured JSON. We are building it now and onboarding a small early access group first.
Get early access
Paste any of the prompts below into ChatGPT, Claude, or Cursor with your API key in your environment. The AI reads our OpenAPI spec, generates working code, and runs it.
Build a Python script using the ASINSpotlight Scraping API. Spec: https://www.asinspotlight.com/scraping-api-docs/openapi.json API base URL: https://api.asinspotlight.com (note: api., not www.) Auth: x-api-key header from env var ASINSPOTLIGHT_API_KEY Endpoint: GET /v1/product?asin=…&marketplace=… For a list of ASINs, fetch prices across us, uk, tr. Save asin, marketplace, bb_price, in_stock to SQLite. Run as a daily cron. Stop early if meta.usage.requests_remaining drops below 50. Cap parallelism at 5.
Build a Python keyword tracker using the ASINSpotlight Scraping API. Spec: https://www.asinspotlight.com/scraping-api-docs/openapi.json API base URL: https://api.asinspotlight.com (note: api., not www.) Auth: x-api-key header from env var ASINSPOTLIGHT_API_KEY Endpoint: GET /v1/search?keyword=…&marketplace=… For each keyword across us and uk, take the first 20 entries from data.shallow_parts and save to SQLite. Tell me which ASINs entered or fell off since yesterday. Run as a daily cron. Don't make a follow-up /v1/product call per result — shallow_parts already has price, rating, reviews, monthly demand.
Build a Python Buy Box monitor using the ASINSpotlight Scraping API. Spec: https://www.asinspotlight.com/scraping-api-docs/openapi.json API base URL: https://api.asinspotlight.com (note: api., not www.) Auth: x-api-key header from env var ASINSPOTLIGHT_API_KEY Endpoint: GET /v1/offers?asin=…&marketplace=… Watch a list of 20 ASINs every hour. When a new seller appears in product_sellers_info with price below mine, send me a Telegram message with the seller name and price. Use APScheduler. The offer with is_pinned: true is the current Buy Box owner, so read ownership from that flag rather than inferring it from price. Don't conflate /v1/product (returns the Buy Box price) with /v1/offers (returns the seller list). You need /v1/offers here. A single /v1/offers response is one page of the offer panel, not the whole seller list. Read data.total_pages and fetch page=2..total_pages, concatenating product_sellers_info across pages, before deciding a seller is new. Buy Box ownership itself only needs page 1, since is_pinned appears there and nowhere else.
Why this works
Clean REST + API key, self-describing JSON, full OpenAPI spec — every property of an API your AI integrates well. The same things that make the docs readable for you make them readable for it.
| Capability | ASINSpotlight | Rainforest | ScraperAPI / Oxylabs | Keepa | Amazon PA-API | DIY scraper |
|---|---|---|---|---|---|---|
| Real-time data | raw HTML | cached, hours-days old | depends | |||
| Structured JSON (no parsing) | you parse HTML | |||||
| BSR + full seller list per ASIN | extract yourself | cached | extract yourself | |||
| All Amazon marketplaces | per-region work | |||||
| AI / LLM-friendly docs + OpenAPI | n/a | |||||
| Battle-tested scraping engine | n/a | starts from zero | ||||
| Price per 1k requests | $0.49 | $1.50 | $1.50–7.35 | subscription | free | proxy + dev cost |
Strong product, well-documented, established. 20–30× more expensive at typical scale, and US-first when it comes to marketplace coverage. If your traffic is low enough that the bill doesn't matter and you don't sell cross-border, Rainforest is a fine choice.
Raw-HTML proxy services. The hard part — turning HTML into structured product data — stays your problem. The parser breaks every time Amazon changes a div. You're trading a known cost (us) for a hidden engineering cost (you).
Built for historical analysis (price/BSR charts over months). Data is cached, hours to days old. If your product needs current sellers, prices, or stock, it's the wrong category — different tool, not a competitor.
Officially supported, free with an Associates account. Missing the data that actually matters: no BSR, no full seller list, no offers panel, strict rate limits, requires an Associates account in good standing. Useful for a narrow product-info display, not for serious tooling.
Amazon blocks naive scrapers within 20–30 requests. Proxies, layout-change detection, marketplace quirks, reliability at scale — you'd be running a scraping-infrastructure team. We already do.
Stackable packs. No subscription, no auto-renew. Buy more whenever — packs combine.
The same scraping engine has been powering ASINSpotlight's desktop app for years across thousands of paying users. The API isn't a side project — it's the same engine, exposed as a service. Status page, transparent incident reporting, support that responds in hours not days.
That's our problem, not yours. The team has spent years keeping up with layout changes, rate limits, and per-marketplace quirks. Your code keeps calling the same endpoint with the same shape. When Amazon shifts something, you don't refactor.
Standard REST. JSON responses. API key in a header. No proprietary SDK to install, no OAuth dance, no vendor-specific data model. Your integration code is yours — the field names are descriptive enough that you could swap the endpoint base URL and probably keep going.
Yes, by design. Predictable JSON shape, self-descriptive field names (no codes that need lookup tables), complete OpenAPI spec, working code examples, error messages an LLM can read and act on. The AI-integration guide gives the model everything it needs to generate working code from a one-line prompt.
The demo at the top of this page is real — it hits the live API and returns the same JSON your code will receive. No signup. The free quota covers enough requests to validate that the data is what you need before you ever pay.
Your data layer just works, doesn't break overnight, and isn't on your roadmap.
Engineering time goes to your product, not to maintaining an Amazon parser.
Your AI generates working code in one session, not a sprint.
Automation runs on schedule whether you're at the laptop or not.
