Best Buy Product Scraper

Public HTTP search — default mode

Search Best Buy's public US catalog without a provider API key or source-script execution. Send searchQuery and maxItems; public-http is the default dataSource. maxPages (1–10, default5) bounds source pages, and requestTimeoutSecs (5–45) bounds each request. A configured fixed US residential proxy is used without direct fallback or identity rotation. Robots/access failures stop the run. Local execution uses apify run; local storage is not uploaded.

The initial server-rendered page may populate only two primary cards. The Actor reads JSON data from the source's Apollo transport without evaluating JavaScript, joins detailed products by observed name and canonical URL, and validates the native query and page-number echo before using cp pagination. Placeholders, unrelated carousels, duplicate URLs, query/page mismatches and unknown fields cannot pad output. A shorter source result is reported as partial. maxItems is an upper bound, not a guarantee that a source has that many products.

Output

Native fields include title, URL, observed SKU and product code, customer/regular price, source savings, rating/review count, brand/model/condition, images, seller and structured fulfillment where provided. Genuine zero/false values survive. Native price.skuId must match the product SKU. A generic Available status is not inferred. Shipping/pickup facts apply only to the returned destination ZIP/store context. Individual review text is not replaced by ratings. Unknown prices and specifications are omitted rather than guessed.

Every row points to its saved source response and native event/page. RUN_SUMMARY records page observations, result counts and coverage limits. One native GraphQL probe of the public client endpoint timed out; this HTTP runtime does not retry that endpoint. It extracts only fields already present in the public page response.

Other source modes

The pre-existing website and official-api choices remain explicit alternatives. They are not certified by public-http search verification. Select another dataSource for productUrl mode or individual review requirements. The official Products API requires BESTBUY_API_KEY; its individual-review and modern nonnumeric direct-URL coverage remains unverified. Store credentials only in encrypted Actor environment variables.

Example

Input: {"searchQuery":"laptop","maxItems":10,"maxPages":5,"includeReviews":false,"dataSource":"public-http"}

Run apify validate-schema and apify run. Inspect the dataset and source-linked RUN_SUMMARY. Production must be checked independently against the exact uploaded source, Actor ID and ScrapeAI owner; schema/offline checks are not ten-record live proof. No fixtures or cached records are used as live results.