MercadoLibre Products, Reviews, Q&A and Sellers

Scrape public MercadoLibre and MercadoLivre search listings, product pages, reviews, buyer questions, and seller storefronts across 18 Latin American marketplaces. Use the actor for regional price monitoring, competitive research, product discovery, seller analysis, promotion tracking, e-commerce automation, and datasets for AI workflows.

The actor keeps the established four-mode contract: search, product, seller, and reviews. It collects public MercadoLibre information and returns the results through the Dataset using the documented input and output fields.

What this actor extracts

Depending on the selected mode and what the public page exposes, records can include:

• Product titles, URLs, thumbnails, images, brands, descriptions, conditions, SKU values, attributes, variations, and breadcrumbs.
• Current and previous prices, discounts, currency, installments, availability, shipping, returns, and promotions.
• Ratings, rating counts, review counts, inline reviews, buyer questions, and seller answers.
• Seller IDs, names, reputation, power-seller status, official-store details, and storefront metadata.
• Seller profiles, catalog items, featured items, menu corridors, banners, logos, and public storefront descriptions.
• Stable identifiers such as publicationId, productId, catalogProductId, userProductId, variationId, sellerId, and officialStoreId.

Availability depends on the marketplace, category, listing, country, page type, and current public HTML. Missing values are returned as null where the schema allows it; arrays and nested objects remain in the same flat dataset record rather than becoming separate records.

Use cases

Price, discount, and promotion monitoring

Schedule the same search across countries and compare currentPrice, previousPrice, discount, installments, and promotions. Keep scrapedAt, countryCode, resultPosition, and productUrl to build a historical series.

Competitive research and share of search

Use searchSort, resultPosition, resultType, brand, sellerName, sellerReputation, and isOfficialStore to study product visibility and seller competition. Run once with sponsored listings excluded and again with them included when you need the commercial view.

Product research and demand signals

Search across marketplaces and inspect titles, ratings, review counts, shipping signals, availability, categories, and seller context. Narrow the keyword or split countries into separate runs when you need comparable regional samples.

Catalog enrichment and matching

Search mode can add product fields to selected listing records. Product and catalog identifiers help relate multiple listings to the same product family, while product URLs and seller IDs support matching across exports.

Seller and storefront analysis

Seller mode emits a storefront profile when enabled, catalog rows, and optional featured-item rows. Use it for brand monitoring, assortment analysis, seller discovery, and public storefront comparisons.

Automation and AI datasets

Send the Dataset to spreadsheets, databases, BI tools, alerting systems, RAG pipelines, or AI agents. Use the Apify API, official clients, webhooks, schedules, or MCP to connect the actor to an existing workflow.

Quick start

1. Open karamelo/mercadolibre-scraper-products-reviews-qa-sellers in Apify.
2. Choose Scraping Mode.
3. Enter a Search Query for Search mode, or provide product/seller URLs for the URL-based modes.
4. Select the Target Marketplace and, for Search mode, a Search Sort Order.
5. Set Max Search Pages for Search mode, or Max Seller Pages for Seller mode. These limits are independent.
6. Turn on Scrape Product Details only when enrichment is needed. Max Search Items then controls how many search rows receive additional product fields; it does not cap the listing records collected from each page.
7. In Reviews mode, use Max Reviews to cap the total review records emitted by the run.
8. Start the run and open the Dataset when it finishes.

Input configuration

The Console starts with the ungrouped fields in this order: Scraping Mode, Search Query, Target Marketplace, Search Sort Order, Max Search Pages, and Include Sponsored Listings. Product URLs is inside Product mode options, followed by the product-detail switches. The remaining fields are grouped into Search mode options, Product mode options, Seller mode options, Advanced, and Reviews mode options.

Field	Type	Group	Description
mode	string	Visible	Select search, product, seller, or reviews; the dropdown labels are Search mode, Product URLs mode, Seller mode, and Reviews mode.
searchQuery	string	Visible	Keyword used to build a Search mode URL.
country	string	Visible	Two-letter marketplace code used for the domain, site ID, language context, and currency.
searchSort	string	Visible	Search order: relevance, price_asc, or price_desc.
maxSearchPages	integer	Visible	Search pages to inspect, from 1 to 20. Search is page-bounded rather than item-bounded.
includePromoted	boolean	Visible	Include sponsored or promoted cards in Search mode.
scrapeProductDetails	boolean	Search mode options	Add available product fields to selected search rows.
maxSearchItems	integer	Search mode options	When enrichment is enabled, maximum search products to open for detail enrichment, from 1 to 5,000. It does not limit listing rows collected from a page.
productUrls	string array	Product mode options	Product URLs used by Product and Reviews modes, up to 100. The Console supports Add, Bulk edit, and text-file upload with one URL per line.
includeReviews	boolean	Product mode options	Keep inline reviews in Product records, or enable review-record output in Reviews mode.
includeQuestions	boolean	Product mode options	Keep buyer questions and seller answers in Product records.
includeVariations	boolean	Product mode options	Keep public color, size, storage, or other product variations.
sellerUrls	string array	Seller mode options	Seller storefront or seller-listing URLs used by Seller mode, up to 50.
sellerMaxPages	integer	Seller mode options	Seller storefront pages to inspect, from 1 to 20. This is independent from maxSearchPages / Max Search Pages.
includeSellerProfile	boolean	Seller mode options	Emit one storefront profile record before seller catalog rows.
includeFeaturedItems	boolean	Seller mode options	Emit an additional featured-item record when public storefront cards are available.
startUrls	string array	Advanced	Direct search, product, review, or seller URLs that supplement the mode-specific fields.
maxConcurrency	integer	Advanced	Maximum parallel product or seller work items, from 1 to 8.
reviewRating	string	Reviews mode options	Selected star rating for Reviews mode: all, 1, 2, 3, 4, or 5.
reviewOrder	string	Reviews mode options	Review ordering: relevance or dateCreated.
maxReviews	integer	Reviews mode options	Maximum total review records emitted in Reviews mode after filtering, from 1 to 5,000.

Input defaults

Field	Default
mode	search
searchQuery	zapatillas
country	AR
searchSort	relevance
maxSearchPages	1
includePromoted	false
scrapeProductDetails	false
maxSearchItems	5
productUrls	["https://www.mercadolibre.com.ar/apple-iphone-16-128-gb-negro-distribuidor-autorizado/p/MLA1040287808"]
includeReviews	true
includeQuestions	true
includeVariations	true
sellerUrls	["https://listado.mercadolibre.com.ar/tienda/xiaomi"]
sellerMaxPages	1
includeSellerProfile	true
includeFeaturedItems	false
startUrls	[]
maxConcurrency	8
reviewRating	all
reviewOrder	relevance
maxReviews	15

The Advanced group contains the optional startUrls and maxConcurrency controls. Product URL uploads use the same documented URL list as Add and Bulk edit entries.

Search limits and enrichment behavior

Search mode processes up to maxSearchPages pages. If scrapeProductDetails is off, Max Search Items is not used. If enrichment is on, maxSearchItems determines how many listing records receive additional product fields. The remaining listings from the requested pages are still returned as search records.

When enrichment is off, Search mode emits listing-level fields only. When enrichment is on, each successfully enriched selected listing is emitted in place of its original search row and marked with mode: "product". If no additional fields are available, that row remains a listing-only mode: "search" record. The enrichment cap does not limit the page-bounded search listing rows.

Product mode returns separate publication and catalog identifiers, seller identity and reputation, power-seller status, official-store status, stock, condition, ratings, prices, shipping, promotions, category/domain, and product attributes when available. Seller mode processes up to sellerMaxPages pages per seller URL and keeps this page limit separate from Search mode's maxSearchPages. Reviews mode emits at most maxReviews review records across the run after applying reviewRating and reviewOrder; the default is 15. Neither mode is controlled by the Search mode enrichment count.

Supported marketplaces

Use the two-letter code in country. The actor maps it to the correct MercadoLibre or MercadoLivre domain, site ID, and currency.

Country	Code	Domain	Currency
Argentina	AR	mercadolibre.com.ar	ARS
Bolivia	BO	mercadolibre.com.bo	BOB
Brazil	BR	mercadolivre.com.br	BRL
Chile	CL	mercadolibre.cl	CLP
Colombia	CO	mercadolibre.com.co	COP
Costa Rica	CR	mercadolibre.co.cr	CRC
Dominican Republic	DO	mercadolibre.com.do	DOP
Ecuador	EC	mercadolibre.com.ec	USD
Guatemala	GT	mercadolibre.com.gt	GTQ
Honduras	HN	mercadolibre.com.hn	HNL
Mexico	MX	mercadolibre.com.mx	MXN
Nicaragua	NI	mercadolibre.com.ni	NIO
Panama	PA	mercadolibre.com.pa	USD
Paraguay	PY	mercadolibre.com.py	PYG
Peru	PE	mercadolibre.com.pe	PEN
El Salvador	SV	mercadolibre.com.sv	USD
Uruguay	UY	mercadolibre.com.uy	UYU
Venezuela	VE	mercadolibre.com.ve	VES

For regional comparisons, run one task per country so currency, domain, language, ranking, and public commercial conditions remain explicit.

Scraping modes

Search mode

Enter a keyword and select a marketplace. Search results are processed page by page. Choose the order with searchSort:

• relevance: MercadoLibre's default order;
• price_asc: lowest price first;
• price_desc: highest price first.

Each unique listing record can include title, brand, thumbnail, current and previous price, discount, rating, shipping, seller context, identifiers, ranking, and search context. Sponsored cards are excluded by default. When enrichment is enabled, selected rows are returned with mode: "product" and can also include product details such as description, condition, images, attributes, variations, questions, reviews, returns, and seller data when available.

Product mode

Provide one or more product URLs. Product mode emits a full product record per successfully processed URL, including title, description, brand, condition, SKU, image gallery, prices, discount, currency, stock, ratings, installments, shipping, returns, seller identity and reputation, official-store status, attributes, variations, inline reviews, questions, and breadcrumbs when available.

Reviews mode

Provide product URLs. Reviews mode emits one flat record per public review available for each supplied product URL. The product identifier in each URL is used as the review target, so Reviews mode does not emit a Product record. Use reviewRating to select a star rating, reviewOrder to request relevance order or newest-first order, and maxReviews to cap the total review records emitted by the run. The default maximum is 15; deeper review coverage depends on what the public marketplace response serves.

Seller mode

Provide seller storefront or seller-listing URLs; the default is https://listado.mercadolibre.com.ar/tienda/xiaomi. Seller mode follows up to sellerMaxPages pages per seller URL and can emit one profile record per seller, catalog item rows, and optional featured corridor rows. Profile records can include the storefront name, identifiers, followers, public description, images, and menu links when available.

Input examples

Search with price sorting and enrichment

{
  "mode": "search",
  "searchQuery": "zapatillas deportivas",
  "country": "AR",
  "searchSort": "price_asc",
  "maxSearchPages": 3,
  "includePromoted": false,
  "scrapeProductDetails": true,
  "maxSearchItems": 5
}

This returns listing records from all three selected pages. At most five of those rows receive additional product fields.

Product details

{
  "mode": "product",
  "country": "MX",
  "productUrls": [
    "https://www.mercadolibre.com.mx/apple-iphone-17-pro-256-gb-azul-profundo/p/MLM55308781"
  ],
  "includeReviews": true,
  "includeQuestions": true,
  "includeVariations": true
}

Reviews with a rating filter

{
  "mode": "reviews",
  "country": "MX",
  "productUrls": [
    "https://www.mercadolibre.com.mx/apple-iphone-17-pro-256-gb-azul-profundo/p/MLM55308781"
  ],
  "reviewRating": "4",
  "reviewOrder": "dateCreated",
  "maxReviews": 25
}

Seller storefront

{
  "mode": "seller",
  "country": "AR",
  "sellerUrls": [
    "https://listado.mercadolibre.com.ar/tienda/xiaomi/"
  ],
  "sellerMaxPages": 2,
  "includeSellerProfile": true,
  "includeFeaturedItems": true
}

Output dataset

Every record includes mode and scrapedAt. Search, product, and review records also include their marketplace and product context when available. Dataset results are shown directly in the default display with the declared fields.

Search rows use mode: "search". If enrichment is enabled, the selected rows are returned in their original position with mode: "product"; their available search values are retained and the product fields are added where available.

Listing and product fields

Search records use this exact public structure: mode, thumbnailUrl, currency, currentPrice, previousPrice, discount, title, installments, promotions, soldQuantity, freeShipping, searchKeyword, rating, reviewCount, productUrl, shipping, highlights, countryCode, siteId, resultPosition, totalResults, publicationId, catalogProductId, userProductId, variationId, isPricePerQuantity, isInternationalPurchase, shipsFrom, inStock, sellerName, isOfficialStore, officialStoreName, sku, resultType, isPromoted, availableIn, brand, categoryId, domainId, sellerId, vertical, verticalId, location, and scrapedAt. Search prices and totalResults are numeric when available; discount is the public label such as "6% OFF"; unavailable optional values are null or empty strings according to the field type.

Product records can additionally include images, description, availability, inStock, condition, breadcrumbs, attributes, variations, reviews, questions, returnPolicy, sellerReputation, sellerPowerStatus, and relevant vehicle fields. On Product pages, reviewCount and ratingCount use the public total count when the page exposes it; Search listing counts may differ because they describe the listing context.

Review and seller fields

Reviews mode uses reviewId, reviewRating, reviewText, reviewDate, reviewCountry, and reviewLikes. Seller records can include sellerSlug, sellerProfileUrl, sellerListingUrl, shopId, shopName, shopType, officialStoreId, totalFollowers, bannerUrl, logoUrl, storefrontUrl, sellerDescription, menuCorridors, corridorName, and corridorUrl when available.

Example search record

{
  "mode": "search",
  "thumbnailUrl": "https://http2.mlstatic.com/D_NQ_NP_979958-MLA105250865291_012026-O.webp",
  "currency": "CLP",
  "currentPrice": 939990,
  "previousPrice": 999990,
  "discount": "6% OFF",
  "title": "Apple iPhone 17 (256 GB) - Azul neblina - Distribuidor Autorizado",
  "installments": "12 cuotas de 78332 sin interés",
  "promotions": "",
  "soldQuantity": 500,
  "freeShipping": true,
  "searchKeyword": "iphone",
  "rating": 4.9,
  "reviewCount": 0,
  "productUrl": "https://www.mercadolibre.cl/apple-iphone-17-256-gb-azul-neblina-distribuidor-autorizado/p/MLC1055308852",
  "shipping": "Llega gratis hoy Enviado por FULL",
  "highlights": "APPLE TIENDA OFICIAL",
  "countryCode": "CL",
  "siteId": "MLC",
  "resultPosition": 1,
  "totalResults": 3949,
  "publicationId": "MLC1694275125",
  "catalogProductId": "MLC1055308852",
  "userProductId": "MLCU3409642961",
  "variationId": "MLC1055308852",
  "isPricePerQuantity": false,
  "isInternationalPurchase": false,
  "shipsFrom": "",
  "inStock": null,
  "sellerName": "Apple Tienda oficial",
  "isOfficialStore": true,
  "officialStoreName": "Apple Tienda oficial",
  "sku": "MLC1055308852",
  "resultType": "ORGANIC",
  "isPromoted": false,
  "availableIn": "",
  "brand": "",
  "categoryId": "MLC1055",
  "domainId": "MLC-CELLPHONES",
  "sellerId": "",
  "vertical": "CORE",
  "verticalId": "CORE",
  "location": "",
  "scrapedAt": "2026-08-20T00:00:00.000Z"
}

Arrays and nested objects remain inside each flat Dataset record.

Complete output property dictionary

Group	Fields	Use
Product	title, productUrl, thumbnailUrl, images, brand, description, condition, sku	Product discovery and catalog enrichment.
Price	currency, currentPrice, previousPrice, discount, installments	Regional price and discount history.
Search context	resultPosition, resultType, isPromoted, searchKeyword, totalResults, soldQuantity	Result ranking and search analysis.
Shipping	freeShipping, shipping, shippingDetails, shipsFrom, availability, inStock, returnPolicy	Delivery, availability, and returns analysis.
Reviews and Q&A	rating, ratingCount, reviewCount, reviews, questions, reviewRating, reviewText, reviewDate	Product quality and review monitoring.
Seller	sellerId, sellerName, sellerReputation, sellerPowerStatus, isOfficialStore, officialStoreName	Seller comparison and official-store analysis.
Taxonomy	breadcrumbs, categoryId, domainId, vertical, verticalId, location, kind	Segmentation by category, domain, and record type.
Identifiers	siteId, publicationId, productId, catalogProductId, userProductId, variationId, sku	Relationships and historical comparison.
Seller profile	sellerSlug, sellerProfileUrl, sellerListingUrl, shopId, shopName, shopType, officialStoreId, totalFollowers, bannerUrl, logoUrl, storefrontUrl, sellerDescription, menuCorridors, corridorName, corridorUrl	Public storefront information.
Audit	mode, countryCode, scrapedAt	Record type, marketplace, and collection time.

Pricing

This Actor uses pay-per-event pricing. Platform usage is included.

Output record	Charge event	Price
Search result	search_item	$2.00 per 1,000 records
Product result, including Search enrichment	product_detail	$6.00 per 1,000 records
Review result	review_item	$1.00 per 1,000 records
Seller profile, listing, or featured result	seller_item	$2.00 per 1,000 records

Every saved result record is charged once according to the result type above. Search enrichment records are charged as product results when they are emitted with mode: "product"; review and seller records are charged as their respective result types. Seller catalog records are search-style results limited to one seller, so seller_item uses the same rate as search_item.

Apify API

Replace APIFY_TOKEN with your Apify token. Use the actor name exactly as published: karamelo/mercadolibre-scraper-products-reviews-qa-sellers.

JavaScript / Node.js

import { ApifyClient } from 'apify-client';

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

const run = await client.actor(
  'karamelo/mercadolibre-scraper-products-reviews-qa-sellers',
).call({
  mode: 'search',
  searchQuery: 'zapatillas deportivas',
  country: 'AR',
  searchSort: 'price_asc',
  maxSearchPages: 3,
  includePromoted: false,
  scrapeProductDetails: true,
  maxSearchItems: 5,
});

const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);

Install the client with:

$npm install apify-client

Python

import os
from apify_client import ApifyClient

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

run = client.actor(
    'karamelo/mercadolibre-scraper-products-reviews-qa-sellers'
).call(run_input={
    'mode': 'search',
    'searchQuery': 'zapatillas deportivas',
    'country': 'AR',
    'searchSort': 'price_asc',
    'maxSearchPages': 3,
    'includePromoted': False,
})

items = list(client.dataset(run['defaultDatasetId']).iterate_items())
print(items)

Install the client with:

$pip install apify-client

cURL: synchronous dataset response

curl -X POST \
  "https://api.apify.com/v2/acts/karamelo~mercadolibre-scraper-products-reviews-qa-sellers/run-sync-get-dataset-items?format=json&clean=true" \
  -H "Authorization: Bearer APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "mode": "search",
    "searchQuery": "zapatillas deportivas",
    "country": "AR",
    "searchSort": "price_asc",
    "maxSearchPages": 3,
    "includePromoted": false
  }'

cURL: asynchronous run

curl -X POST \
  "https://api.apify.com/v2/acts/karamelo~mercadolibre-scraper-products-reviews-qa-sellers/runs" \
  -H "Authorization: Bearer APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "mode": "search",
    "searchQuery": "zapatillas deportivas",
    "country": "AR",
    "searchSort": "price_asc",
    "maxSearchPages": 3,
    "includePromoted": false
  }'

After the run completes, use its defaultDatasetId to retrieve JSON, CSV, or another Dataset export format supported by Apify.

MCP and AI agents

Apify MCP can expose this actor to compatible clients such as ChatGPT, Claude, Cursor, Codex, and other MCP-enabled tools. Complete Apify authorization when the client requests it.

{
  "mcpServers": {
    "apify": {
      "url": "https://mcp.apify.com?tools=karamelo/mercadolibre-scraper-products-reviews-qa-sellers"
    }
  }
}

Example agent prompt:

Run karamelo/mercadolibre-scraper-products-reviews-qa-sellers for
"notebook gamer" in Mexico (MX), inspect 3 search pages, sort by lowest price,
exclude sponsored listings, and return title, current price, currency, discount,
shipping, rating, seller, and product URL.

Example JavaScript context preparation:

const contextForAgent = items.map((item) => ({
  mode: item.mode,
  title: item.title,
  currentPrice: item.currentPrice,
  currency: item.currency,
  discount: item.discount,
  freeShipping: item.freeShipping,
  rating: item.rating,
  seller: item.sellerName,
  productUrl: item.productUrl,
}));

console.log(JSON.stringify(contextForAgent));

Automation and integrations

• Schedules: run the same query hourly, daily, or weekly to build history.
• Webhooks: notify another service when a run succeeds or fails.
• Make and Zapier: connect datasets to spreadsheets, databases, email, or Slack.
• Google Sheets and Excel: build price, promotion, and seller dashboards from Dataset exports.
• Data warehouses: load JSON or JSONL into BigQuery, Snowflake, PostgreSQL, or another pipeline.
• AI agents: connect through MCP or consume the Dataset with JavaScript or Python.

Consistent-results recommendations

• Start with one or two pages to validate the query and fields.
• Use includePromoted: false for organic-ranking analysis and true for the full commercial view.
• Keep countryCode, searchSort, resultPosition, productUrl, and scrapedAt in historical comparisons.
• Interpret currentPrice together with currency; do not mix ARS, BRL, MXN, and other currencies without conversion.
• Use product, listing, and variation identifiers according to the entity you want to compare.
• Check nullable fields before calculating averages or sending records to production.
• Use a reasonable page count and schedule frequency for your data need.

Limitations

• Search returns public results from the pages processed by maxSearchPages; it does not promise the marketplace's full total.
• Search enrichment applies to maxSearchItems selected listing rows. Listing rows beyond that count are still returned with their search fields.
• Descriptions, questions, reviews, attributes, availability, shipping, and stock depend on what MercadoLibre publishes for each product.
• Price, sold-count, result totals, shipping, and delivery dates can be rounded, summarized, or absent.
• Category, location, availability, page design, and experiments can change the fields visible in a response.
• A page can produce fewer records when the marketplace hides, duplicates, removes, or blocks results.
• Product and seller pages are public-page targets; no private orders, account history, or unpublished seller data are collected.

Frequently asked questions

Which MercadoLibre countries are supported?

All 18 configured marketplaces are supported: AR, BO, BR, CL, CO, CR, DO, EC, GT, HN, MX, NI, PA, PE, PY, SV, UY, and VE. Use the two-letter code in country.

Can I compare one product across countries?

Yes. Run a separate task for each country, then compare the resulting Datasets while retaining currency, countryCode, scrapedAt, and productUrl.

Can I sort results by price?

Yes. Set searchSort to price_asc or price_desc.

Does maxSearchItems limit the number of search listings?

No. Search is page-bounded by maxSearchPages. When scrapeProductDetails is enabled, maxSearchItems controls how many listing rows receive additional product fields. It does not remove later listing rows from the Dataset.

Can I exclude sponsored listings?

Yes. Set includePromoted to false. Set it to true when sponsored cards should remain in the search output.

Do I need to provide login credentials or cookies?

No. The actor is designed for public pages and does not accept or require target-site credentials or cookies.

How do I create a price monitor?

Keep mode, searchQuery, country, searchSort, maxSearchPages, and includePromoted consistent across scheduled runs. Compare currentPrice, previousPrice, discount, promotions, resultPosition, and scrapedAt by product or listing identifier.

Can I download the results as CSV or Excel?

Yes. Export the Apify Dataset from the Console or use the Dataset items API with a supported format such as CSV.

Can I use this actor as a product API?

Yes. Run it through the Apify REST API, official JavaScript and Python clients, or MCP. Product and seller URL modes are useful when the input is already known; Search mode is useful when the input is a keyword.

Responsible use and legal notice

Collect only public information and use it according to applicable terms, privacy rules, and marketplace policies. Avoid unnecessary personal data and excessive request loads. This Karamelo actor is independent and is not affiliated with, sponsored by, or endorsed by MercadoLibre, MercadoLivre, or Apify.

Support

For troubleshooting, share the search query, country, mode, page count, search sort, sponsored-listing setting, run ID, and an example of the unexpected field. Never share tokens or other secrets.