Pangolinfo

First thing in the morning, Amazon sellers check whether a rival cut the price overnight. Across a few hundred products in several countries, that check runs until lunch. The usual next step, a homemade program that copies the pages, needs a supply of borrowed internet addresses and repairs each time Amazon changes its layout. Pangolinfo's Amazon Scraper API loads the pages itself and returns them as tidy data fields.

A request names a marketplace and what to fetch, for instance a search keyword or a product by its ASIN, Amazon's product number. Thirteen Amazon marketplaces are covered, Japan and Brazil included. An optional postal code helps because price and delivery dates change with the buyer's location. Slow jobs can report back later through a webhook, which sends the result to a web address on your side.

The parser that turns Amazon pages into fields is on GitHub under the MIT licence, while the collection runs as a hosted service. AI agents reach the data through the Amazon Data MCP server. For teams without a developer, AMZ Data Tracker runs the same collection on a schedule and writes the results to Excel or Lark Base. The schedule can be set anywhere between hourly and monthly.