Mercado Livre Brazil Scraper

Search the current mercadolivre offers catalog and export structured Brazilian marketplace product records. Mercado Livre Brazil Scraper collects public titles, sellers, prices, discounts, availability, ratings, sales labels, images, listing IDs, product URLs, and source-query context from mercadolivre.com.br.

Use it for scheduled catalog snapshots, competitor-price monitoring, seller-offer review, deal discovery, and data pipelines that need typed JSON instead of marketplace pages.

What does Mercado Livre Brazil Scraper do?

The Actor reads Mercado Livre Brazil's public server-rendered offers pages. It can:

• match one or more product keywords against current offer titles and seller names;
• collect a specific Mercado Livre /ofertas page;
• keep products inside a BRL price range;
• filter products by public seller or official-store name;
• deduplicate records by Mercado Livre listing ID;
• stop at a user-defined result limit;
• produce one clean default-dataset record per accepted listing.

The Actor is optimized for repeat snapshots. Schedule the same input and compare records by listingId to identify price, seller, discount, rating, or availability changes.

Who is it for?

E-commerce analysts can build a current offer-price benchmark.

Brands and manufacturers can review public marketplace positioning and seller visibility.

Retail and procurement teams can find products inside a budget band.

Marketplace agencies can export offer cards for client reporting.

Data engineers can feed normalized Mercado Livre product rows into a warehouse, dashboard, spreadsheet, webhook, or monitoring job.

Developers can use the Actor as a hosted, schedulable alternative to maintaining HTML extraction and Apify integrations themselves.

Why use this Actor?

• Brazil-specific source: the implementation targets mercadolivre.com.br and BRL prices.
• Monitoring-ready IDs: listingId provides a stable key for comparing snapshots.
• Useful seller context: public seller names are included when Mercado Livre displays them.
• Input parity: query, seller, and price filters apply to explicit offers-page URLs too.
• Fail-closed behavior: traffic-verification or unrecognized pages cause a failed run instead of misleading empty success.
• No automatic residential proxy: the current route uses efficient direct HTTP and does not expose an unmeasured costly fallback.
• Integration-friendly output: primary records go to the default Apify dataset.

What Mercado Livre product data is extracted?

Field	Type	Meaning
listingId	string	Stable listing ID, normally an MLB… value
productId	string or null	Catalog product ID exposed by the URL
title	string	Public listing title
seller	string or null	Public seller or official-store name
price	number	Current price in BRL
originalPrice	number or null	Previous price when shown
discountPercent	number or null	Displayed discount percentage
currency	string	Always BRL for this Actor
rating	number or null	Displayed rating from zero to five
soldQuantityText	string or null	Source-formatted public sales label
availability	string	available when present in the current offers catalog
imageUrl	string or null	Public product thumbnail URL
url	string	Public product URL without tracking fragment
sourceQuery	string or null	First matching input keyword
matchedQueries	string[]	Every matching input keyword
sourceUrl	string	Offers page that produced the record
position	integer	One-based card position on that source page
scrapedAt	string	ISO 8601 extraction timestamp

Nullable values reflect what Mercado Livre displays on each card. A missing rating or previous price is returned as null, not fabricated.

How to scrape Mercado Livre Brazil offers

1. Open the Actor input page.
2. Enter one or more product keywords, such as notebook.
3. Optionally add a supported Mercado Livre /ofertas URL.
4. Add a seller substring or BRL price range when needed.
5. Set the maximum-products limit to control dataset size and total budget.
6. Click Start.
7. Open the Products output or export the default dataset as JSON, CSV, Excel, XML, or RSS.
8. For monitoring, schedule the Task and compare snapshots by listingId.

Start with a small result limit while validating a new workflow.

Input parameters

Input	Required	Default	Description
queries	conditionally	notebook	Up to 20 case-insensitive keywords matched against titles and sellers
startUrls	conditionally	offers page in prefill	Mercado Livre Brazil URLs under /ofertas
seller	no	—	Seller-name substring filter
minPrice	no	—	Minimum accepted current price in BRL
maxPrice	no	—	Maximum accepted current price in BRL
maxItems	no	100	Maximum unique rows, from 1 to 5,000

Provide at least one non-empty query or one start URL. Unsupported hosts and paths fail validation before collection.

Example keyword search:

{
  "queries": ["notebook"],
  "maxItems": 10
}

Example monitoring input:

{
  "queries": ["smartphone", "camera"],
  "minPrice": 100,
  "maxPrice": 1500,
  "maxItems": 20
}

Example source-page export:

{
  "queries": [],
  "startUrls": [
    { "url": "https://www.mercadolivre.com.br/ofertas" }
  ],
  "maxItems": 20
}

Output example

A current run returns records shaped like this:

{
  "listingId": "MLB49089309",
  "productId": "MLB49089309",
  "title": "Notebook ASUS Vivobook 15 M1502, AMD Ryzen 7, 8 GB RAM, 512 GB SSD",
  "seller": "Asus",
  "price": 2927,
  "originalPrice": 3999,
  "discountPercent": 26,
  "currency": "BRL",
  "rating": 4.8,
  "soldQuantityText": "+100 vendidos",
  "availability": "available",
  "imageUrl": "https://http2.mlstatic.com/D_NQ_NP_...webp",
  "url": "https://www.mercadolivre.com.br/notebook-asus-vivobook-15/p/MLB49089309",
  "sourceQuery": "notebook",
  "matchedQueries": ["notebook"],
  "sourceUrl": "https://www.mercadolivre.com.br/ofertas",
  "position": 23,
  "scrapedAt": "2026-08-06T20:52:30.000Z"
}

Values change with the live catalog. The schema, field names, and types remain stable.

How much does it cost to monitor Mercado Livre offers?

The Actor uses pay-per-event pricing:

• one start event per run;
• one item event for each validated product saved to the default dataset;
• no item charge for duplicate, filtered-out, invalid, or failed records.

At the BRONZE rate, a run starts at $0.005 and each accepted item is $0.0015. Examples before platform-usage charges:

Accepted products	BRONZE calculation	Approximate total
1	0.005 + (1 × 0.0015)	0.0065 USD
10	0.005 + (10 × 0.0015)	0.0200 USD
25	0.005 + (25 × 0.0015)	0.0425 USD
100	0.005 + (100 × 0.0015)	0.1550 USD

Apify subscription tiers can use different item rates. The Console estimate shown before a run is the authoritative rate for your account tier. Compute and storage usage may also apply according to your Apify plan.

Catalog monitoring workflow

Create an Apify Task with a stable input and schedule it daily or weekly. Export every run to a warehouse table keyed by listingId and scrapedAt. Then compare the latest two snapshots.

Common change rules include:

• price decreased or increased;
• discountPercent appeared or disappeared;
• seller changed;
• a listing did not appear in the latest current-offers snapshot;
• a new listing ID matched a tracked query;
• rating or public sales-volume text changed.

Absence from one offers-page snapshot means the listing was not present on that sampled source page; it does not prove that the product is unavailable across the entire marketplace.

Export and integration patterns

Use Apify dataset exports for one-time analysis:

• JSON for applications and event pipelines;
• CSV or Excel for analysts and agencies;
• XML or RSS for legacy workflows;
• Google Sheets through an Apify integration;
• webhooks to trigger downstream processing after a successful run.

For repeat monitoring, store listingId, price, seller, sourceUrl, and scrapedAt. Those fields provide the most useful snapshot identity and provenance.

Run through the Apify API with cURL

Replace APIFY_TOKEN with your token:

curl -X POST \
  "https://api.apify.com/v2/acts/automation-lab~mercado-livre-brazil-scraper/runs?token=APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"queries":["notebook"],"maxItems":10}'

To wait for completion and return dataset rows:

curl -X POST \
  "https://api.apify.com/v2/acts/automation-lab~mercado-livre-brazil-scraper/run-sync-get-dataset-items?token=APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"queries":["notebook"],"maxItems":10}'

Run from JavaScript

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('automation-lab/mercado-livre-brazil-scraper').call({
    queries: ['notebook'],
    maxItems: 10,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);

Run from Python

import os
from apify_client import ApifyClient

client = ApifyClient(os.environ["APIFY_TOKEN"])
run = client.actor("automation-lab/mercado-livre-brazil-scraper").call(
    run_input={"queries": ["notebook"], "maxItems": 10}
)
items = client.dataset(run["defaultDatasetId"]).list_items().items
print(items)

Use with MCP and AI agents

Add the Actor to Claude Code through Apify MCP:

claude mcp add --transport http apify \
  "https://mcp.apify.com?tools=automation-lab/mercado-livre-brazil-scraper"

Claude Desktop, Cursor, and VS Code

Claude Desktop, Cursor, and VS Code can use this HTTP MCP configuration:

{
  "mcpServers": {
    "apify": {
      "url": "https://mcp.apify.com?tools=automation-lab/mercado-livre-brazil-scraper"
    }
  }
}

Example prompts showing MCP usage:

• “Run Mercado Livre Brazil Scraper for notebook offers and return the 10 cheapest current records.”
• “Collect current smartphone and camera offers between R$100 and R$1,500.”
• “Compare these Mercado Livre listing IDs with yesterday's dataset and summarize price changes.”

The AI client still needs an Apify token and permission to run the Actor.

Limits and responsible interpretation

This Actor covers the current public offers catalog and supported /ofertas pages. It does not claim marketplace-wide inventory, authenticated account data, private seller information, complete product details, reviews, coupons, historical prices, or stock counts.

Keyword matching is performed against the current page's product titles and public seller names. A query can legitimately produce zero rows when no current offer card matches it.

Mercado Livre can change HTML, offer composition, or traffic controls. The Actor recognizes the expected page shape and fails on verification shells instead of silently producing an empty dataset.

A successful record means the product card was displayed as an available current offer when scraped. It is not a guarantee that checkout or delivery remains available later.

Reliability and troubleshooting

The run failed with an unrecognized or verification page.

Retry later once. If the source still returns a verification shell, inspect the run log and do not interpret the failed run as an empty catalog.

My query returned zero products.

Open the source offers page and check whether a current card contains the exact keyword in its title or seller. Try a broader term, remove the seller filter, or use the source-page export example.

A URL was rejected.

Only mercadolivre.com.br hosts and paths beginning with /ofertas are accepted. Product-detail, account, and generic search URLs are intentionally outside this Actor's supported source route.

The same item appeared on two source pages.

The Actor deduplicates within a run by listingId, so the first accepted occurrence is retained.

I received fewer rows than maxItems.

maxItems is a ceiling, not a promised count. Filters, deduplication, and current source-page inventory can all reduce output.

Legal and responsible use

The Actor extracts public marketplace offer-card data. Use it only for lawful purposes and in accordance with Mercado Livre's terms, applicable database and copyright rules, privacy laws, and your contractual obligations.

Do not use the output to profile private individuals, circumvent access controls, automate purchases, impersonate sellers, or send unsolicited messages. Avoid collecting more data or running more frequently than your legitimate business purpose requires.

You are responsible for your input, retention period, downstream processing, and compliance decisions. This documentation is operational guidance, not legal advice.

Related Automation Lab actors

• Amazon Product Search Results Scraper for comparable Amazon marketplace search snapshots.
• Amazon Bestsellers Scraper for category bestseller monitoring.
• Amazon Sellers Scraper for Amazon seller-focused workflows.

These Actors target different sources. They are useful for multi-market analysis but do not extend Mercado Livre coverage inside this Actor.

FAQ

Does it use the official Mercado Livre API?

No. The anonymous official search endpoint is not available to this runtime. The Actor parses public server-rendered offer cards and records the source URL for provenance.

Does it scrape all Mercado Livre products?

No. It samples the current public offers catalog or the explicit supported offers pages you provide. It does not claim complete marketplace inventory.

Can I monitor prices over time?

Yes. Schedule a stable input and compare runs by listingId. Keep scrapedAt and sourceUrl so every observed value retains time and source context.

Are filtered-out records charged?

No. The item event is emitted only after a record passes validation, query matching, deduplication, seller filtering, price filtering, and the result limit.

Can I provide product-detail or generic search URLs?

No. The Actor rejects unsupported paths rather than pretending they are covered. Use /ofertas sources or keywords matched against the current default offers catalog.

Does it need a proxy configuration?

No user proxy input is exposed. The working implementation uses direct HTTP. This keeps the runtime lightweight and prevents hidden residential-proxy cost.