The eBay product API is the item-id-keyed detail endpoint: hand it item ids and it returns each listing as one structured row — title, subtitle, price with its parsed value and currency, condition, the seller and their positive-feedback percent, the availability line, photos, the item-specifics grid and the category breadcrumb.
Each id is fetched over the TLS tier without a browser and parsed into a flat row. The item-specifics grid — brand, model, size and whatever else the seller filled in — comes back as specifics, a label-to-value object, so the attributes that vary by category are preserved rather than flattened away. Ended or removed listings return under failed and are never billed.
$0.003 per delivered product, up to 20 products per run. Nothing delivered means nothing charged.
You post an items array. Each entry can be a bare item id or a full /itm/ URL — both normalise to the same fetch, so ids from your own records or from a discovery run drop straight in.
Every item becomes one detail fetch parsed into a flat row. Three behaviours matter before you wire it in:
price preserves the displayed string such as US $98.00, while price_value and currency split out the number and the currency for maths and cross-market comparison.seller and seller_feedback_pct come back together, so reputation is data you filter on rather than something you eyeball.failed, not as a half-empty row that would skew a price average later.Need the ids first? Run a keyword through the eBay search collector, which returns item ids and /itm/ URLs this endpoint accepts. For the Amazon equivalent, the Amazon product API keys off ASINs the same way.
The only required input is items, an array of up to 20 item ids or /itm/ URLs. max_results caps how many the run fetches, a cheap guard when the array is built by another job. country selects the eBay marketplace and the matching residential exit, since the same item can price differently across sites.
Each item is one flat row. price_value is the number behind price, paired with a best-effort currency. condition is the listing's own condition text, and availability preserves the quantity or availability line as shown. images holds up to six photo URLs, categories the breadcrumb path, and specifics the label-to-value attributes that differ from one category to the next.
$0.003 per delivered product ($3 per 1,000). Nothing delivered means nothing charged, and the $2 monthly free credit covers roughly 666 products here. Volume tiers take up to 30% off.
$0.003 per delivered product, $3 per 1,000. A 5,000-item watchlist refreshed daily is
Item ids that no longer resolve are returned under failed and are not billed, so an ageing id list costs coverage, not money. The free monthly allowance of $2 covers roughly 666 products, and GET /v1/scraper/collectors reports the price your key pays.
eBay's Browse API is the sanctioned way to read item data, and it returns clean structured JSON. The cost is the plumbing: you register a developer account, create an application, run the OAuth client-credentials flow to mint and refresh tokens, stay inside per-application call limits, and get your app approved for production before the higher limits unlock. For a small enrichment job that is a lot of ceremony.
This collector reads the same public item page a shopper sees, so there is no application, no OAuth token to rotate and no marketplace scope to request — your QuantumProxies key is the only credential. You trade the API's guaranteed field catalogue for the page's public fields, including the seller feedback percent and the availability line exactly as rendered, which is what most pricing and reputation work needs.
Fetching one eBay item page is trivial. Fetching them reliably at volume is the whole problem:
Handing it over buys a published schema and a bill that only moves on delivered rows. If you would rather keep your own fetcher and just want clean exits under it, the residential proxies are the network this collector runs on.
Refresh a set of item ids and diff price_value and availability to spot under-priced listings and sold-through demand across marketplaces.
Read seller and seller_feedback_pct across a watchlist to filter out low-trust sellers before you buy, source or benchmark against them.
Turn a list of item ids into structured records using specifics, condition, categories and images, without anyone re-keying a listing by hand.
One run takes up to 20 item ids; longer lists are simply more runs, which parallelise up to your plan's rate limit — 60 requests a minute on pay-as-you-go, up to 1,200 on the top tier. images is capped at six photos per item, and ended or removed listings come back under failed at no charge.
Legally: item listings are shown publicly to every visitor, and collecting public data is generally lawful in most jurisdictions, but eBay's terms restrict automated access, seller feedback is personal data in some readings, and your downstream use carries its own duties. This is not legal advice — take advice for your specific case.
eBay's Browse API is free to call but requires a registered developer app, OAuth tokens and production approval. This collector needs none of that, and it comes with a $2 monthly allowance, roughly 666 products at $0.003 each, before any charge.
No. Because the data is read from the public item page rather than eBay's Browse API, there is no developer registration, application or OAuth token to maintain. Your QuantumProxies key is the only credential.
Run a keyword through the eBay search collector, which returns item ids and /itm/ URLs, then pass either straight into this endpoint. The search and detail collectors are designed to be chained.
Yes. The item-specifics grid comes back as a label-to-value object in specifics, and the seller's name and positive-feedback percent are returned as their own fields, alongside condition, availability and the category breadcrumb.