Costco Scraper — Prices, Item Numbers & Ratings

Costco Scraper is an Apify Actor for Costco.com public catalog data. It accepts a keyword, a Costco search or category URL, or a direct product URL, and returns one normalized record per product with title, price, item number, brand, rating, reviews, availability, image, and source details. It is usable through Apify MCP, and each record represents one Costco product from the live catalog context.

Best fit and connected workflows

This Actor fits workflows that start with Costco.com search, category, or product pages and end with structured product rows in an Apify dataset. Common routing patterns include:

• Keyword discovery from Costco search results such as laptop or olive oil.
• Category monitoring from Costco listing pages.
• Direct product lookups from .product.<id>.html URLs.
• AI agent workflows through Apify MCP when an agent needs Costco product facts, provenance, and a bounded result count.

For broader pipelines, the dataset can feed spreadsheets, databases, dashboards, or monitoring jobs that compare Costco product facts over time.

Practical scenario

Maya, a procurement analyst, starts with the keyword robot vacuum and a modest maxResults value. The Actor returns product rows with title, public price, item number, rating, review count, availability, product URL, and image URL. Maya uses those fields to shortlist a few Costco items, then opens the product URLs for a closer review and compares the item numbers against her internal buying notes.

Input fields

Field	Type	Purpose
searchQuery	string	Keyword search on Costco.com, such as laptop or patio set.
startUrls	array of strings	Costco search, category, or product URLs. Product URLs return one full detail record.
maxResults	integer	Maximum number of product records to return across all sources.
minPrice	integer	Keep products at or above this USD price.
maxPrice	integer	Keep products at or below this USD price.
minRating	integer	Keep products at or above this star rating.
enrichProductDetails	boolean	Open product detail pages for breadcrumbs, full specs, description, and image gallery.
proxyConfiguration	object	Optional proxy settings for the browser fallback.

Focused JSON input example

{
  "searchQuery": "robot vacuum",
  "maxResults": 10,
  "minPrice": 100,
  "minRating": 4,
  "enrichProductDetails": true,
  "proxyConfiguration": {
    "useApifyProxy": true
  }
}

Output fields

The default dataset contains product records. The live dataset contract includes the following fields.

Field	Type	Meaning
recordType	string	Dataset row type, with product as the product record value.
title	string	Costco product title shown in the catalog or product page.
productUrl	string	Canonical Costco product URL for the record.
itemId	string	Costco item ID when available.
itemNumber	string	Costco item number when available.
sku	string	Costco SKU when available.
manufacturerSku	string	Manufacturer SKU when available.
brand	string	Brand name extracted from the product data.
price	number	Current price in the listed currency.
originalPrice	number	Original price when shown.
savings	number	Savings amount when shown.
savingsPercent	integer	Savings percentage when shown.
currency	string	Currency code such as USD when present.
requiresMembership	boolean	Membership pricing signal from Costco.
rating	number	Average star rating when present.
reviewCount	integer	Public review count when present.
availability	string	Public stock or availability signal.
imageUrl	string	Main product image URL.
images	array	Additional image URLs from the gallery.
breadcrumbs	array	Category breadcrumbs from detail enrichment.
category	string	Category name when available.
subCategory	string	Subcategory name when available.
dimensions	string	Product dimensions when available.
weight	string	Product weight when available.
material	string	Material when available.
color	string	Color when available.
description	string	Product description when enriched.
features	array	Feature bullets when available.
specs	object	Structured specifications when enriched.
shippingInfo	string	Shipping information when available.
freeShipping	boolean	Free shipping flag when available.
promoFlag	string	Promotion label when available.
variantCount	integer	Number of variants when available.
position	integer	Result position in a listing.
searchQuery	string	Search term used for the run.
site	string	Site source marker when available.
sourceUrl	string	Source page URL.
retrievalMethod	string	Route used to obtain the record.
scrapedAt	string	Timestamp of collection.
_warnings	array	Record-level warnings when present.

Illustrative dataset record

{
  "recordType": "product",
  "title": "Example Costco Product",
  "productUrl": "https://www.costco.com/example.product.100000001.html",
  "itemId": "100000001",
  "itemNumber": "100000001",
  "brand": "Example Brand",
  "price": 49.99,
  "originalPrice": 59.99,
  "savings": 10,
  "savingsPercent": 17,
  "currency": "USD",
  "requiresMembership": false,
  "rating": 4.7,
  "reviewCount": 218,
  "availability": "In Stock",
  "imageUrl": "https://cdn.example.com/product.jpg",
  "searchQuery": "example",
  "sourceUrl": "https://www.costco.com/CatalogSearch?keyword=example",
  "retrievalMethod": "listing",
  "scrapedAt": "2026-07-20T12:00:00.000Z"
}

How it works

The Actor supports keyword search and Costco URLs. Listing pages are auto-detected and paginated until maxResults is reached. Product URLs return one full detail record. When enrichProductDetails is enabled for keyword or listing runs, the Actor opens each product detail page to add the item number, brand, breadcrumbs, full specifications, description, and image gallery.

Costco.com is protected by Akamai. The Actor first uses a managed US JavaScript-unlocker route, then falls back to browser-proxy handling when needed. The default dataset contains the product rows, and the key-value store contains machine-readable OUTPUT and RUN_SUMMARY records for the run outcome.

Pricing

This Actor uses Pay per event plus Apify platform usage.

• Actor start event: charged once per run start.
• Product event: charged for each product record written to the dataset.

For example, a run that returns ten product records uses eleven Actor events total: one start event and ten product events. Apify compute and proxy usage are billed separately by the platform. For current pricing details, open the live Pricing tab for this Actor when planning a run size or schedule.

Use with AI agents (MCP)

This Actor is available through Apify MCP as a structured Costco product-data tool. The exact Actor identity is khadinakbar/costco-scraper.

A practical MCP use case is a retrieval step where an agent gathers Costco catalog facts, then reads the dataset for product rows and the run store for the outcome metadata.

Search Costco for wireless headphones, return up to 10 product rows, and then summarize the title, public price, item number, rating, review count, availability, and source URL for each product. Read the dataset for product rows and the OUTPUT record for the run outcome.

The agent can interpret the dataset as structured product facts and use sourceUrl plus scrapedAt as provenance for downstream comparisons. For listing runs, position and searchQuery help preserve search context. For pagination and cost guidance, maxResults sets the product cap across all sources, and each product written to the dataset is a billable event. The OUTPUT and RUN_SUMMARY records provide outcome, route, validation, and billing diagnostics when present.

Run with the Apify API

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });

const run = await client.actor('khadinakbar/costco-scraper').call({
  searchQuery: 'wireless headphones',
  maxResults: 5,
  enrichProductDetails: true,
});

const { items } = await client.dataset(run.defaultDatasetId).listItems();
for (const item of items) {
  console.log(item.title, item.price, item.itemNumber);
}

from apify_client import ApifyClient
import os

client = ApifyClient(os.environ["APIFY_TOKEN"])

run = client.actor("khadinakbar/costco-scraper").call(
    run_input={
        "searchQuery": "protein powder",
        "maxResults": 5,
        "enrichProductDetails": False,
    }
)

for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item.get("title"), item.get("price"), item.get("itemNumber"))

Best results and outcome guidance

Start with a narrow keyword or a specific Costco URL and a modest maxResults value. Use startUrls when you already have a search, category, or product page. Turn on enrichProductDetails when you need breadcrumbs, full specs, description, or the image gallery. Use minPrice, maxPrice, and minRating to shape the returned dataset before downstream processing. For AI agent workflows, read both the dataset and the OUTPUT record so the agent can separate product facts from run outcome metadata.

Continue the workflow

• Then use Coupang Product Scraper - Prices & Listings to extend Costco Scraper — Prices, Item Numbers & Ratings with a neighboring commerce research source when the brief calls for Coupang data.
• Then use Avito Scraper — Listings, Prices, Sellers & Item Data to extend Costco Scraper — Prices, Item Numbers & Ratings with a neighboring commerce research source when the brief calls for Avito data.

Design note

I found that the live dataset contract makes recordType and scrapedAt required on every product row. That means every stored product record carries a clear record type and a collection timestamp.

FAQ

Can I start from a Costco search page instead of a keyword?

Yes. startUrls accepts Costco search pages, category pages, and product pages. The Actor auto-detects the URL type.

When should I use searchQuery?

Use searchQuery when you want the Actor to open Costco search results from a keyword such as laptop or patio set.

What happens with a Costco product URL?

A product URL returns one full detail record. This is useful when you already know the exact product page and want the full structured record.

Can I keep the output focused on a price range or rating range?

Yes. Use minPrice, maxPrice, and minRating to shape the returned product rows before they are written to the dataset.

Is this Actor usable through Apify MCP?

Yes. It is an Apify Actor usable through Apify MCP, and it returns structured Costco product data that an agent can read from the dataset.

Where are run outcome details stored?

The default dataset stores the product rows. The key-value store contains OUTPUT and RUN_SUMMARY records with outcome and execution diagnostics.

Responsible use

Use this Actor only for public Costco.com catalog pages and publicly visible product data. Keep requests within the requested scope, respect applicable law and site terms, and use the returned source URL and timestamp when you cite or compare product facts.