MercadoLibre Scraper

MercadoLibre Scraper

Extract product data, buyer reviews, Q&A, and seller storefront profiles from MercadoLibre — the largest e-commerce marketplace in Latin America — across all 9 country sites, no coding required.

Pull a MercadoLibre product's full spec sheet, buyer Q&A, and top reviews in one call, mine every review across 9 marketplaces with a deep rating/sort sweep, or fetch a seller's storefront profile — followers, brand info, and corridor menu included.

Copy to your AI assistant

Copy this block into ChatGPT, Claude, Cursor, or any LLM to start using this actor.

Use the Apify actor "parsebird/mercadolibre-scraper" (call it via ApifyClient: client.actor("parsebird/mercadolibre-scraper").call(run_input={...})). It has three modes selected via the "mode" input field: "reviews" (default) returns paginated buyer reviews for one or more "productUrls" across MercadoLibre's 9 LATAM sites, using a deep-sweep strategy across rating and sort buckets to recover up to ~950 reviews per product; "product" returns full product-detail fields for "productUrls" — title, brand, price, installments, currency, availability, condition, sku, rating, seller (id/name/reputation/power status/official store), up to 50+ attributes, breadcrumbs, variations, up to 20 buyer Q&A pairs, and the top 5 reviews; "seller" returns one storefront-profile row per "sellerUrls" (must be a /tienda/, /loja/, or /pagina/ URL) with followers, brand info, banner, corridor menu, and compliance flags. Other key inputs: "country" (MX/AR/BR/EC/DO/CL/CO/PE/UY, default MX), "maxItems" (default 50), "reviewRating" (default "all"), "reviewOrder" (default "relevance"), "includeReviews"/"includeQuestions"/"includeVariations" (product mode, default true). Output is pushed to the default dataset — one row per review/product/seller. Full input/output schema and examples: https://apify.com/parsebird/mercadolibre-scraper/input-schema and https://apify.com/parsebird/mercadolibre-scraper/api. Get your API token at https://console.apify.com/settings/integrations.

What does MercadoLibre Scraper do?

MercadoLibre Scraper is a MercadoLibre data extraction tool covering three focused modes in one actor:

• 🛒 Product mode — full PDP data: title, brand, price, original price and discount %, installments, currency, availability, condition, SKU, rating and review counts, seller identity and reputation, official-store flag, 50+ product attributes, category breadcrumbs, color/size/storage variations, up to 20 buyer questions with seller answers, and the top 5 reviews.
• ⭐ Reviews mode — full paginated buyer reviews for one or more products. When the rating filter is left at "all", the actor runs a deep sweep — 12 concurrent queries across every star rating and both sort orders — to recover far more unique reviews than a single query can (MeLi caps a single query at ~300 offset / ~330 reviews; the deep sweep can recover up to ~950 on a popular product).
• 🏪 Seller mode — one storefront-profile row per seller: follower count, creation date, tags, banner and logo URLs, free-trial dates, brand registry info, official-store status, and the full corridor/category menu.
• 🌎 All 9 LATAM marketplaces — Mexico, Argentina, Brazil, Chile, Colombia, Peru, Uruguay, Ecuador, and the Dominican Republic, selected with a single country input.
• ⚡ Runs on the Apify platform: schedule recurring runs, trigger via API or webhook, and export results as JSON, CSV, or Excel.

Q&A extraction and the multi-bucket review-recovery strategy make this actor useful for use cases most MercadoLibre scrapers don't cover well — pulling buyer-seller conversations and maximizing review coverage on popular products.

What data can you extract from MercadoLibre?

Field	Description
title, brand, sku	Product identity
price, originalPrice, discountPercentage, currency	Pricing
installments	Interest-free installment plan (quantity, per-payment amount, total)
availability, condition	Stock status and item condition
rating, ratingCount, reviewCount	Aggregate rating data
sellerId, sellerName, sellerReputation, sellerPowerStatus, isOfficialStore	Seller identity and reputation
attributes	Product specs (RAM, dimensions, color, model, and 50+ more, varies by category)
variations	Color/size/storage options, when the listing has them
questions	Buyer questions with seller answers (up to 20 per product)
reviews	Top 5 reviews (product mode) or full paginated review text, rating, date, likes, and media (reviews mode)
totalFollowers, bannerUrl, menuCorridors, freeTrialStart/End	Seller storefront profile (seller mode)

See the Output tab for the complete field list per mode.

How to scrape MercadoLibre with this actor

1. Open the actor's Input tab in Apify Console, or use the API.
2. Pick a Mode: reviews to mine buyer reviews, product for full product-page data, or seller for a storefront profile.
3. Paste one or more MercadoLibre URLs into Product URLs (reviews/product mode) or Seller URLs (seller mode). Pasting into Start URLs also works — the mode and country are auto-detected from the URL.
4. Pick the Country matching the marketplace, set Max Items, and click Start.
5. Download results as JSON, CSV, or Excel from the Dataset tab, or pull them via the API / Integrations (Zapier, Make, Google Sheets, and more).

Output example

Reviews mode, one row per review:

{
  "mode": "reviews",
  "country": "MX",
  "productId": "MLM27172669",
  "catalogProductId": "MLM27172678",
  "reviewId": 803375213,
  "reviewRating": 5,
  "reviewText": "Producto chino pero 100% original...",
  "reviewDate": "2023-11-21T16:00:16Z",
  "reviewDateText": "Hace más de 1 año",
  "reviewCountry": "México",
  "reviewLikes": 2537,
  "reviewMedia": ["https://http2.mlstatic.com/D_NQ_NP_698307-MLA72919050762_112023-F.jpg"],
  "scrapedAt": "2026-08-16T07:08:12.148Z"
}

Product mode, one row per product:

{
  "mode": "product",
  "country": "MX",
  "title": "Localizador Xiaomi Tag Tracker con Bluetooth blanco",
  "url": "https://www.mercadolibre.com.mx/localizador-xiaomi-tag-tracker-con-bluetooth-blanco/p/MLM65678975",
  "brand": "Xiaomi",
  "price": 295,
  "originalPrice": 685.25,
  "discountPercentage": 56,
  "currency": "MXN",
  "installments": { "quantity": 24, "amount": 17.83, "total": 427.92 },
  "rating": 4.8,
  "ratingCount": 11805,
  "sellerId": 454333952,
  "sellerName": "Xiaomi",
  "sellerReputation": "5_green",
  "sellerPowerStatus": "platinum",
  "isOfficialStore": true,
  "attributes": [
    { "name": "Marca", "value": "Xiaomi" },
    { "name": "Alcance máximo", "value": "10 m" }
  ],
  "questions": [
    { "text": "Es recargable?", "answer": "No es recargable, funciona con una pila botón CR2032...", "answerDate": "15/08/2026" }
  ],
  "reviews": [
    { "rating": 5, "text": "¡Producto muy bueno! Funciona perfectamente..." }
  ],
  "scrapedAt": "2026-08-16T08:11:10.203Z"
}

Use cases

• Price and stock monitoring — track a product's price, discount, installments, and stock status over time.
• Review mining — pull every review for a product across all 9 marketplaces for sentiment analysis or product-feedback dashboards.
• Competitive Q&A research — see what buyers ask sellers and how well (and how fast) they answer.
• Seller due diligence — check a storefront's reputation tier, official-store status, follower growth, and compliance flags before partnering or comparing competitors.
• Category research — collect specs and attributes across a catalog of products for a comparison table or feed.

How it works

1. Reviews mode calls MercadoLibre's internal reviews API directly (noindex/catalog/reviews/...) — no browser or proxy needed, so it's fast and inexpensive. With the rating filter left at "all", the actor fans out 12 concurrent requests (5 star ratings + "all", × 2 sort orders) and de-duplicates by review ID to maximize unique reviews recovered.
2. Seller mode fetches the storefront's /tienda/ or /loja/ page anonymously and parses the embedded page-state JSON for profile fields — also no proxy needed.
3. Product mode renders the page with a real, fingerprinted browser session (Apify's residential proxy, routed to the selected country) because MercadoLibre's product pages sit behind a JavaScript-executing anti-bot challenge that plain HTTP requests cannot pass. The actor extracts the same internal page-state JSON MercadoLibre's own frontend uses, plus the page's schema.org Product JSON-LD, then fetches the listing's dedicated Q&A page for the full buyer question list.
4. Every mode pushes normalized rows straight to the Apify dataset, ready to export or pull via API.

Input parameters

Parameter	Type	Required	Default	Description
mode	string	No	reviews	reviews, product, or seller
country	string	No	MX	MercadoLibre marketplace (MX, AR, BR, EC, DO, CL, CO, PE, UY)
productUrls	array	For reviews/product	—	MercadoLibre product URLs (catalog /p/MLM… or article MLM-…)
maxItems	integer	No	50	Max items to return (reviews mode: per product)
reviewRating	string	No	all	Filter reviews mode by star rating, or all for the deep sweep
reviewOrder	string	No	relevance	relevance or dateCreated
sellerUrls	array	For seller mode	—	Storefront URLs (/tienda/, /loja/, or /pagina/)
includeReviews	boolean	No	true	Product mode: include the top 5 reviews
includeQuestions	boolean	No	true	Product mode: include buyer Q&A
includeVariations	boolean	No	true	Product mode: include variations
startUrls	array	No	—	Direct URLs; mode/country auto-detected
maxConcurrency	integer	No	8	Parallel requests in reviews mode's deep sweep
useResidentialProxy	boolean	No	true	Informational for product mode, which always uses residential proxy

See the Input tab for full descriptions and enum values.

How much does it cost to scrape MercadoLibre?

This actor uses pay-per-event pricing — you pay only for records actually pushed to the dataset, not for compute time.

Plan	Price per event	Price per 1,000
Free	$0.0025	$2.50
Bronze	$0.0017	$1.70
Silver	$0.0012	$1.20
Gold	$0.0010	$1.00

Every pushed row — one review, one product, or one seller profile — counts as one result-scraped event. Paid Apify plans (Bronze, Silver, Gold) get a lower per-event rate automatically. A run that pulls 300 reviews for one product costs about $0.75 on the free plan (as low as $0.30 on Gold); a single product-mode fetch costs $0.0025. Apify's free trial covers enough runs to try every mode before committing to a plan. Product mode additionally consumes Apify's residential proxy bandwidth (~$8/GB, billed separately by Apify) since it requires a real browser session to get past MercadoLibre's anti-bot check.

Is it legal to scrape MercadoLibre?

Scraping publicly available data, like MercadoLibre product listings and reviews, is generally legal — see Apify's blog post on the legality of web scraping for a fuller discussion. This actor only extracts publicly visible page data; it does not bypass logins, paywalls, or access non-public information. You are responsible for how you use the collected data and for complying with MercadoLibre's own terms of service.

Other scrapers by ParseBird

• eBay Sold Listings Scraper — sold-listing prices and history from eBay
• 1688 Wholesale Scraper — wholesale product data from 1688.com
• Shopify Products Scraper — product catalogs from any Shopify store
• MercadoLibre Scraper (Español) — the same tool, with a Spanish-language interface

FAQ

Why isn't there a search or category-listing mode? MercadoLibre's search and category-listing pages (listado.*) sit behind a stricter, apparently behavior-based anti-bot layer than product pages, and a reliable bypass isn't in place yet. Product, reviews, and seller-profile modes are all fully working and use verified extraction against live MercadoLibre data.

Why does seller mode return only one row per store instead of a full catalog? For the same reason — a seller's item catalog is served on listado.* pages, which this actor doesn't currently scrape reliably. Seller mode returns the storefront's profile data (followers, brand, corridor menu, compliance flags), which comes from the /tienda/ or /loja/ page and isn't affected by that block.

Does reviews mode really cover all 9 marketplaces? Yes — pass the matching country for MX, AR, BR, CL, CO, PE, UY, EC, or DO, and the actor builds the review-API URL for that site's domain and site ID.

Why is seller mode not supported for Argentina? Argentina's storefront pages (mercadolibre.com.ar/tienda/...) use a different rendering framework than the other 8 sites, so the profile-page parser doesn't apply there yet.

Can I run this on a schedule or trigger it via API? Yes — use Apify's Scheduler for recurring runs, or call the actor via the API using Python, JavaScript, or any HTTP client. See the code snippets below.

Something not working, or want a feature added? Open an issue on the actor's Issues tab in Apify Console — feedback and bug reports are welcome.

API usage examples

Python (apify-client):

from apify_client import ApifyClient

client = ApifyClient("<YOUR_API_TOKEN>")
run = client.actor("parsebird/mercadolibre-scraper").call(run_input={
    "mode": "reviews",
    "country": "MX",
    "productUrls": ["https://www.mercadolibre.com.mx/apple-iphone-15-256-gb-negro/p/MLM27172669"],
    "maxItems": 100,
})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["reviewText"])

JavaScript (apify-client):

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: '<YOUR_API_TOKEN>' });
const run = await client.actor('parsebird/mercadolibre-scraper').call({
    mode: 'product',
    country: 'MX',
    productUrls: ['https://www.mercadolibre.com.mx/apple-iphone-15-256-gb-negro/p/MLM27172669'],
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items[0].title, items[0].price);

Legal and compliance

This actor extracts publicly available MercadoLibre page data for research, monitoring, and analysis purposes. It does not access private accounts, bypass authentication, or store personal buyer information beyond what MercadoLibre displays publicly (e.g. a reviewer's country or a seller's public storefront details). Use of this actor and the data it collects is your responsibility — review MercadoLibre's Terms of Service before large-scale or commercial use.