Mercado Libre Scraper — LATAM Prices, Listings & Specs

Disclaimer: Unofficial tool — not affiliated with, sponsored by, or endorsed by Mercado Libre or MercadoLibre, Inc or its affiliates. Data is read from publicly accessible pages only. No login. You are responsible for complying with applicable law (including GDPR where personal data appears) and the site’s terms. No warranty on accuracy or availability. Provided for informational and research use.

Extract structured product listings and full specifications from Mercado Libre across 18 Latin American country sites — Argentina, Brazil, Mexico, Colombia, Chile, Peru, and more. Search by keyword, category URL, seller storefront, or direct product link. Collect titles, prices, discounts, installments, seller info, shipping badges, ratings, sold counts, catalog IDs, technical attributes, stock, and image galleries.

Runs over country-targeted residential proxies. Keyword search fetches visible listado HTML over HTTP first (Chrome TLS; last retry uses a Googlebot UA), paginates with _Desde_N URLs, then batch-enriches seller/category via /items when that API answers. The public search API is OAuth-gated and is skipped on 403. Camoufox or Playwright Firefox is a fallback when listado serves a JS interstitial. Search returns the SERP as shown unless you turn on strictTitleMatch. If the run only recovers a handful of Ofertas cards after a wall, it fails instead of succeeding with 3–10% of maxItems. A run_coverage row records requested vs delivered.

Best for: Mercado Libre price monitoring, LATAM e-commerce market research, competitor catalog tracking, resale arbitrage, product specification enrichment, and bulk JSON/CSV export pipelines.

Crawloop Marketplace & E-commerce Suite — LATAM Mercado Libre plus Amazon, AliExpress, and Shopee search.

Mercado Libre Listings Scraper ◄── you are here	Amazon Search Scraper	AliExpress Search Scraper	Shopee Search Scraper
LATAM listings, PDP specs, 18 country sites	Amazon keyword SERP, ASIN, badges	AliExpress retail search, sold count	Shopee regional search

Also see: Naver Search Scraper · eBay Sold Listings Scraper

---

When to use this Actor

• Mercado Libre price tracking — Monitor listing prices, discounts, and installments across Argentina (MLA), Brazil (MLB), Mexico (MLM), Chile (MLC), Colombia (MCO), and 13 more LATAM sites.
• Product catalog enrichment — Enable includeProductDetail to fetch full PDP data: description, spec tables, stock, sold count, variations, and high-resolution images.
• Cross-border market research — Compare the same product keyword across multiple Mercado Libre country sites with local currency output.
• Official store monitoring — Set officialStoresOnly, or paste a /tienda/, /loja/, /pagina/, or _CustId_ URL. Branded /tienda/ pages are landings, not card grids — without cookies the Actor follows catalog links, then matching public offers if listado is blocked.
• Category & filter scraping — Paste any filtered Mercado Libre search or category URL into Start URLs to replicate browser filters exactly.
• Bulk data pipelines — Export structured JSON or CSV datasets for spreadsheets, databases, and BI tools.

When not to use this Actor

• Private seller messaging or account actions — Reads public listing and search pages only; no login, cart, or messaging access.
• Real-time sub-second monitoring — Optimized for reliable bulk extraction, not millisecond-level price tick feeds.
• Non-Mercado Libre sites — Built specifically for Mercado Libre domains across LATAM; not a generic web scraper.
• Niche scoring, TAM, or market-size reports — this Actor exports listing rows only. It does not compute opportunity scores or demand models.
• Treating a thin run as a full catalog — a login/captcha wall plus two cards is a failed scrape. Check the run_coverage row (requested, delivered, blocked).

---

Key Features

• 18 LATAM country sites — MLA, MLB, MLM, MCO, MLC, MPE, MLU, MLV, MEC, MBO, MPA, MPY, MRD, MCR, MNI, MGT, MHN, MSV with automatic country-targeted residential proxy.
• HTTP-first extraction — Visible search cards over residential HTTP. Camoufox / Playwright Firefox only if listado serves a JS interstitial or captcha wall.
• Two extraction modes — Fast search-card mode (title, price, seller, shipping, sold, category) or full PDP enrichment (specs, description, stock, gallery, variations).
• Parallel PDP fetching — Multi-session proxy pool with configurable pdpConcurrency and automatic retry on anti-bot timeouts.
• Rich listing data — Original price, discount percent, installment plans, coupons, condition (new/used/refurbished), free shipping, FULL logistics badge, catalog product ID, seller reputation, sold counts, and ratings.
• Flexible inputs — Keyword search, category URLs, seller storefronts, direct product URLs, price range filters, condition filters, sort (relevance / price / most sold / newest), and official-store-only mode.
• Structured dataset output — Clean JSON rows ready for Apify dataset export, webhooks, and integrations.

---

How it works

1. Hits visible listado HTML over residential HTTP. TLS impersonation rotates Chrome / Firefox / Android profiles; the last attempt uses a Googlebot UA. Extra Sec-Fetch headers are omitted so they do not mismatch the TLS fingerprint. After a few country-sticky walls it tries any residential IP. Pagination follows the on-page next link, or constructed _Desde_49 / _Desde_97 URLs when the pager is missing. The public search API is OAuth-gated (403) and is only used when it answers.
2. If listado HTML is a JS interstitial (account-verification?go=…), Camoufox loads the same search URL, waits for the public challenge to clear, and replays the resulting cookies onto HTTP pagination. Playwright Firefox is the backup crawler: it opens listado first (not the homepage search box) and waits out the same interstitial.
3. On a hard login or captcha wall it may paginate Ofertas, keeping only cards whose titles match the query. A wall plus a thin yield fails the run. Coverage is written to the dataset and COVERAGE in the key-value store.
4. For /tienda/ or /loja/ storefronts, follows on-page catalog links (_Container_, “Ver todos”), then matching public offers if those listado URLs are walled.
5. Parses search-result cards — titles, prices, seller name, shipping, thumbnails, catalog IDs, installments, sold buckets, and ratings — including listings embedded in page JSON when CSS cards are empty.
6. When includeProductDetail is enabled, fetches each product detail page (HTTP first, Playwright if walled). soldQuantityIsEstimate is true when the page shows +N vendidos.
7. Title filtering is off by default (strictTitleMatch: false) so the dataset matches the SERP. Turn it on when you want the Actor to drop off-query cards. Ofertas fallback always title-matches because deals mix unrelated SKUs.

---

Supported country sites

Site ID	Country	Currency	Domain
MLA	Argentina	ARS	mercadolibre.com.ar
MLB	Brazil	BRL	mercadolivre.com.br
MLM	Mexico	MXN	mercadolibre.com.mx
MCO	Colombia	COP	mercadolibre.com.co
MLC	Chile	CLP	mercadolibre.cl
MPE	Peru	PEN	mercadolibre.com.pe
MLU	Uruguay	UYU	mercadolibre.com.uy
MLV	Venezuela	VES	mercadolibre.com.ve
MEC	Ecuador	USD	mercadolibre.com.ec
MBO	Bolivia	BOB	mercadolibre.com.bo
MPA	Panama	USD	mercadolibre.com.pa
MPY	Paraguay	PYG	mercadolibre.com.py
MRD	Dominican Republic	DOP	mercadolibre.com.do
MCR	Costa Rica	CRC	mercadolibre.co.cr
MNI	Nicaragua	NIO	mercadolibre.com.ni
MGT	Guatemala	GTQ	mercadolibre.com.gt
MHN	Honduras	HNL	mercadolibre.com.hn
MSV	El Salvador	USD	mercadolibre.com.sv

---

Input Parameters

Parameter	Type	Default	Description
siteId	String	MLA	Target country site code (see table above).
searchQueries	Array	["iphone"]	Keywords to search (one per line).
categoryUrls	Array	[]	Category or filtered search URLs to crawl.
startUrls	Array	[]	Mix of search, category, seller storefront, or direct product detail URLs.
sellerUrls	Array	[]	Official stores (/tienda/, /loja/), independent shops (/pagina/), profiles, or _CustId_ catalogs. Branded storefronts need cookies for a full listado dump.
priceMin	Integer	—	Minimum price filter in local currency.
priceMax	Integer	—	Maximum price filter in local currency.
condition	String	any	Filter: any, new, used, refurbished.
officialStoresOnly	Boolean	false	Restrict results to Mercado Libre official stores.
sort	String	relevance	Sort: relevance, price_asc, price_desc, sold_desc, newest.
includeProductDetail	Boolean	false	Fetch full PDP: description, specs, stock, variations, gallery.
strictTitleMatch	Boolean	false	Drop cards whose title does not contain the query words. Leave off to export the SERP as shown.
maxItems	Integer	100	Stop after N items (0 = unlimited).
maxPagesPerQuery	Integer	42	Maximum search pages per query (~48 items/page).
maxConcurrency	Integer	2	Parallel search/category crawl jobs. Keep low to reduce captcha walls.
pdpConcurrency	Integer	5	Parallel product detail page fetches (when PDP enabled).
sessionPoolSize	Integer	4	Independent residential proxy sessions (1–10).
cookies / sessionCookies	Array / String	—	Optional logged-in Mercado Libre cookies when search hits a login or captcha wall.
blockAdsTrackers	Boolean	false	Block tracker domains. Leave off unless you are debugging bandwidth.
proxyConfiguration	Object	auto	Apify proxy; residential proxy in target country auto-configured.

Input example — search listings

{
  "siteId": "MLA",
  "searchQueries": ["iphone 15"],
  "maxItems": 500,
  "condition": "new",
  "officialStoresOnly": false,
  "includeProductDetail": false,
  "sort": "price_asc"
}

Input example — official store URL

{
  "siteId": "MLA",
  "sellerUrls": ["https://www.mercadolibre.com.ar/tienda/samsung"],
  "maxItems": 200,
  "includeProductDetail": false
}

A /tienda/ URL is a branded landing. Add cookies / sessionCookies when you need that store's listado rather than public offers that match the brand name.

Input example — full product details (PDP)

{
  "siteId": "MLB",
  "searchQueries": ["samsung galaxy s24"],
  "maxItems": 200,
  "includeProductDetail": true,
  "pdpConcurrency": 8,
  "sessionPoolSize": 4
}

---

Output Format

Each dataset row = one Mercado Libre product listing.

Field	Type	Description
id	String	Listing ID from wid when present (e.g. MLA1787211113).
catalogProductId	String	Catalog / user-product ID from /p/ or /up/ (e.g. MLA55887294).
siteId	String	Country site code (MLA, MLB, MLM, etc.).
title	String	Product title.
permalink	String	Canonical product URL.
price	Number	Current selling price in local currency.
currency	String	Currency code (ARS, BRL, MXN, CLP, etc.).
originalPrice	Number	Price before discount.
discountPercent	Integer	Discount percentage.
installments	Object	Installment quantity, amount, and interest rate.
coupon	String	Visible Mercado Pago / coupon rebate on the search card.
condition	String	new, used, or refurbished.
availableQuantity	Integer	Stock count (PDP mode).
soldQuantity	Integer	Historical sold count from the search card (+N vendidos is a bucket floor).
seller	Object	Id, nickname, official store flag, store name, reputation.
sellerId / sellerName	String	Flattened seller id and nickname for spreadsheet / BI joins.
categoryId / domainId	String	Mercado Libre category and domain ids from the search or items API.
searchKeyword	String	Query that produced the row.
resultPosition	Integer	1-based rank in the result set.
totalResults	Integer	Search API paging.total when the public API answered.
thumbnailUrl	String	Search card thumbnail.
images	Array	Full image gallery (PDP mode).
description	String	Product description text (PDP mode).
gtin	String	GTIN / EAN from PDP JSON-LD or the specs table.
location	Object	Seller city, state, country (PDP mode).
freeShipping	Boolean	Free shipping badge.
fullShipping	Boolean	Mercado Libre FULL logistics delivery.
ratingAverage	Number	Star rating (0–5).
reviewCount	Integer	Total review count when shown on the card or PDP.
isFeatured	Boolean	Oferta del día / deal badge on the search card.
position	Integer	1-based position on the search or offers page.
attributes	Object	Technical specifications key-value pairs (PDP mode).
variations	Array	Color, size, and other variants with IDs and URLs (PDP mode).
scrapedAt	String	Extraction timestamp (UTC ISO 8601).

Output example

{
  "id": "MLA1424728826",
  "catalogProductId": "MLA19597568",
  "siteId": "MLA",
  "title": "Apple iPhone 14 (128 GB) - Blanco Estelar",
  "permalink": "https://www.mercadolibre.com.ar/apple-iphone-14-128-gb-blanco-estelar/p/MLA19597568",
  "price": 1199999,
  "currency": "ARS",
  "originalPrice": 1499999,
  "discountPercent": 20,
  "installments": {
    "quantity": 12,
    "amount": 99999,
    "rate": 0
  },
  "condition": "new",
  "availableQuantity": 15,
  "soldQuantity": 1500,
  "sellerId": "123456",
  "sellerName": "Apple",
  "searchKeyword": "iphone 14",
  "seller": {
    "id": "123456",
    "nickname": "Apple",
    "isOfficialStore": true,
    "storeName": "Apple",
    "reputation": "platinum"
  },
  "freeShipping": true,
  "fullShipping": true,
  "ratingAverage": 4.8,
  "attributes": {
    "Memoria interna": "128 GB",
    "Memoria RAM": "6 GB",
    "Modelo del procesador": "Apple A15 Bionic"
  },
  "scrapedAt": "2026-07-09T10:40:00Z"
}

---

Search mode vs. PDP mode

	Search listings	Product detail (PDP)
Input flag	includeProductDetail: false	includeProductDetail: true
Speed	HTTP listado (browser only if blocked)	Extra request per item
Data depth	Title, price, seller, shipping, sold count, catalog ID, thumbnail	+ description, specs, stock, gallery, variations, GTIN
Best for	Price monitoring, bulk catalog scans	Specification matching, enrichment, inventory analysis

---

Use Cases

Use Case	What you get
LATAM price intelligence	Structured price datasets per keyword across Argentina, Brazil, Mexico, Chile, and more
Competitor monitoring	Track prices, discounts, and installment offers in a niche, including official-store filters
Catalog specification matching	Full technical attribute tables for ERP, PIM, and marketplace listing sync
Resale arbitrage	Compare new vs. refurbished prices and sold quantities to find margin opportunities
Cross-border research	Same keyword on MLA vs. MLB vs. MLM with local currency normalization
Data pipelines	JSON/CSV export or webhook integrations to Sheets, databases, and BI tools

---

Integration examples

Node.js

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('crawloop/mercado-libre-listings-scraper').call({ siteId: 'MLA', searchQueries: ['iphone 15'], maxItems: 50, includeProductDetail: false });
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items.slice(0, 5));

Python

from apify_client import ApifyClient

client = ApifyClient(token)
run = client.actor("crawloop/mercado-libre-listings-scraper").call(
    run_input={"siteId": "MLA", "searchQueries": ["iphone 15"], "maxItems": 50, "includeProductDetail": False}
)
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item.get("id"), item.get("title"), item.get("price"))

cURL

curl "https://api.apify.com/v2/acts/crawloop~mercado-libre-listings-scraper/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"siteId":"MLA","searchQueries":["iphone 15"],"maxItems":50,"includeProductDetail":false}'

MCP and AI assistants

Use this Actor from AI tools via Apify MCP. Connect your Apify account, then call crawloop/mercado-libre-listings-scraper.

Example prompts:

• "Scrape Samsung listings on Mercado Libre Argentina from the official store URL"
• "Run Mercado Libre Listings Scraper for iphone 15 on MLA, max 50, return id, catalogProductId, title, price, currency"
• "Scrape Mercado Libre Mexico for air fryers without PDP detail"

FAQ

Which Mercado Libre countries are supported?

All 18 Mercado Libre regional sites: Argentina, Brazil, Mexico, Colombia, Chile, Peru, Uruguay, Venezuela, Ecuador, Bolivia, Panama, Paraguay, Dominican Republic, Costa Rica, Nicaragua, Guatemala, Honduras, and El Salvador. Set siteId to the corresponding code (e.g. MLA for Argentina, MLB for Brazil).

How does it handle Mercado Libre bot protection?

Keyword search reads visible listado HTML over country residential HTTP (Chrome TLS, then a Googlebot UA if the first profiles wall). Pages after the first use _Desde_N URLs even when the HTML omits the pager. A JS interstitial (account-verification?go=…) is waited out in Camoufox or Playwright Firefox; the public search API is OAuth-gated (403) and is not the primary path. Ofertas is a last-resort, title-filtered fallback — not a full catalog. The Actor does not solve captchas. If a wall is seen and delivered listings are well under maxItems (under half, and under 10 when you asked for 20+), the run fails so pipelines cannot treat two cards as a market. For a store dump behind a hard login wall, add cookies from a logged-in browser and keep blockAdsTrackers off.

Do I need proxies?

Residential proxies in the target country are strongly recommended and auto-configured on Apify Cloud when proxy input is left empty. For PDP mode with high maxItems, increase sessionPoolSize (default 4) to spread load across more IP addresses.

What is the difference between search and PDP mode?

Search mode (includeProductDetail: false) extracts data from search result cards — fast and cost-efficient for price monitoring. Seller nickname and sold counts are frequently missing on cards. PDP mode (includeProductDetail: true) visits each product page for specifications, description, stock, sold quantity, image gallery, and variations. Sold figures shown as +1000 vendidos are bucket floors (soldQuantityIsEstimate: true), not exact sales.

Can I scrape a seller storefront or official store?

Yes. Paste the URL into sellerUrls or startUrls: official stores (/tienda/ or Brazil /loja/), independent shops (/pagina/), seller profiles, or listado _CustId_ catalogs. Official /tienda/ pages are branded landings without listing cards; the Actor follows on-page catalog links. If those listado URLs hit a wall, it returns title-matched public offers — not necessarily every item in that store. Pass session cookies for a fuller catalog. Pagination follows "Siguiente" when present, otherwise constructed _Desde_N listado URLs.

Can I scrape specific categories or advanced filters?

Yes. Open Mercado Libre in your browser, apply filters (category, price range, condition, sort order), copy the URL from the address bar, and paste it into categoryUrls or startUrls. Price and condition filters can also be set directly in the Actor input.

Does it work with used and refurbished listings?

Yes. Set condition to used or refurbished, or leave it as any. The Actor correctly parses user-product listings (/up/MLAU... URLs) alongside catalog products (/p/MLA...).

Can I scrape 1,000+ listings in one run?

Yes. Set maxItems to your target (e.g. 1000). Search mode handles 1,000 items in approximately 3 minutes. PDP mode with pdpConcurrency: 8 and sessionPoolSize: 4 completes 1,000 enriched products in approximately 40 minutes.

Does it export to JSON / CSV?

Yes. Results are written to the Apify dataset. Download JSON, CSV, or Excel from the run, or connect webhooks and integrations. Each run also writes a run_coverage dataset row and a COVERAGE key-value record (requested, delivered, blocked, source).

Why did a run FAIL after returning a few correct cards?

That is intentional. Mercado Libre sometimes serves a login or captcha wall, then Ofertas yields 1–2 title-matched cards. Succeeding in that state makes an automation think it scraped the market. If a wall or Ofertas fallback ran and delivered listings are well under maxItems, the Actor fails. The cards already in the dataset are still valid.

How is this different from a generic web scraper?

This Actor is purpose-built for Mercado Libre: correct URL construction per country, residential-proxy sessions, installment parsing, official store detection, FULL shipping badges, LATAM number formatting, and parallel PDP enrichment with proxy session pooling.

Suite next step

After a LATAM listing dump, compare the same keyword on Amazon Search Scraper or AliExpress Search Scraper.

Related Actors

Actor	Best for
Mercado Libre Listings Scraper ◄── you are here	LATAM Mercado Libre search and PDP listings
Amazon Search Scraper	Amazon keyword SERP, ASIN, badges
AliExpress Search Scraper	AliExpress retail search, sold count
Shopee Search Scraper	Shopee regional search
Naver Search Scraper	Korean 통합검색 rank / Powerlink
eBay Sold Listings Scraper	Completed eBay sold comps