# Oxylabs MCP

> Oxylabs MCP connects your AI client directly to a powerful web scraping suite designed for deep data extraction. It lets you parse complex search engine results pages from Google, Amazon, Bing, and Yandex—whether you need product pricing, article headlines, or raw HTML. Use the universal scraper tool to pull structured data from any website, even those that rely on JavaScript rendering.

## Overview
- **Category:** developer-tools
- **Price:** Free
- **Tags:** web-scraping, proxy-services, data-extraction, serp-parsing, headless-browser

## Description

This MCP gives your AI client the power to scrape and structure data from virtually any corner of the web. Forget building custom scripts just to track competitor pricing or gather market intelligence. You can send your agent to an arbitrary URL using the universal scraper, which handles modern single-page applications that require JavaScript execution. Need competitive SEO data? Use specialized tools like the Google Search tool to pull structured SERP blocks, capturing organic rankings, featured snippets, and ad results—all localized by country. The scope goes deep: you can scrape product listings with ASINs and pricing from Amazon or analyze e-commerce structures on Google Shopping. Integrating this MCP through Vinkius means your AI client connects once and gets access to a massive catalog of data sources, making complex web analysis a conversation away.

## Tools

### scrape_universal
Extracts raw HTML from any arbitrary website, allowing you to enable JavaScript rendering if the site uses a single-page app structure.

### scrape_google_serp
Parses and structures entire Google Search Engine Results Pages, capturing all organic results, ads, and knowledge panels.

### scrape_google_shopping
Extracts structured product listings, including prices and merchant details from Google Shopping search results.

### scrape_google_images
Gathers specific image URLs, dimensions, and source origins by scraping the Google Image Search layout.

### scrape_google_news
Exports time-indexed news articles from Google News blocks, including publisher names and publication dates.

### scrape_amazon_search
Performs structured data extraction by looping over multiple product listings found in an Amazon search result query.

### scrape_amazon_product
Deep-extracts detailed information from individual Amazon product pages, including stock states and variations.

### scrape_bing_serp
Calculates and extracts structured web mapping data from Bing Search Engine Results Pages for comparative analysis.

### scrape_yandex_serp
Performs structural analysis of Yandex search results, useful for market assessment in the CIS/RU region.

### scrape_custom_payload
Dispatches entirely customized JSON payloads to configure advanced proxy types, locations, and deeply nested settings.

## Prompt Examples

**Prompt:** 
```
Search Google for 'best noise cancelling headphones 2024' and show me the top 5 organic results.
```

**Response:** 
```
Top 5 organic results for 'best noise cancelling headphones 2024': 1) Wirecutter — 'The Best Noise-Canceling Headphones' (score: featured snippet). 2) RTINGS — 'The 8 Best Noise Cancelling Headphones'. 3) TechRadar — 'Best noise-cancelling headphones 2024'. 4) Tom's Guide — 'Best ANC headphones'. 5) CNET — 'Best Noise-Canceling Headphones for 2024'. Want me to scrape any of these pages for full content?
```

**Prompt:** 
```
Get the full product details for Amazon ASIN B0C8PSXB2M.
```

**Response:** 
```
Product: Sony WH-1000XM5 Wireless Noise Canceling Headphones. Price: $328.00 (was $399.99 — 18% off). Rating: 4.6/5 (12,847 reviews). Prime: Yes. Colors available: Black, Silver, Midnight Blue. In stock. Key features: 30-hour battery, 8 microphones, touch sensor controls. Sold by Amazon.com. Want me to check pricing on amazon.de or amazon.co.uk?
```

**Prompt:** 
```
Scrape the homepage of techcrunch.com with JavaScript rendering enabled.
```

**Response:** 
```
Successfully scraped techcrunch.com with JS rendering. Page title: 'TechCrunch | Startup and Technology News'. Extracted 42 article headlines, 28 author names, 15 featured images, and 8 category tags. Top headline: 'AI startup raises $200M Series C at $2B valuation'. Content size: 245 KB (after rendering). Want me to extract just the article titles and links in a structured format?
```

## Capabilities

### Analyze Search Engine Results
Parse structured search result blocks from Google (including SERPs, shopping, images, and news) or Bing for immediate market insights.

### Deep E-commerce Data Extraction
Pull detailed product information, pricing history, and seller data across major platforms like Amazon and Google Shopping.

### Universal Web Scraping
Extract raw HTML content from any website URL, enabling JavaScript rendering for modern web pages that require client-side execution.

### International Search Analysis
Gather structured ranking data from niche search engines like Yandex and Bing across different international markets.

## Use Cases

### Tracking competitor price changes in multiple countries
A market analyst needs to know if a major appliance brand is raising prices on Amazon Germany compared to Amazon UK. They ask their agent to run scrape_amazon_search across both regional domains, then use the individual product tool to pull specific ASIN data for comparison.

### Auditing an SEO campaign's performance
An SEO team wants to verify if a new article ranks highly after three weeks. They ask their agent to run scrape_google_serp and compare the featured snippet status and surrounding SERP structure against previous weeks' data.

### Building an image asset database
A content manager needs all visual assets used in top search results. They ask their agent to run scrape_google_images, collecting the necessary URLs and dimensions for a large-scale image library build.

### Collecting broad market news coverage
A PR firm wants to track how often its client is mentioned in major world publications. They ask their agent to run scrape_google_news, gathering time-indexed articles and publisher details from the last quarter.

## Benefits

- You get structured, actionable data instead of messy HTML dumps. Whether using the Google Search tool for SERPs or scrape_amazon_product for deep listings, the output is ready to feed directly into a database.
- Manage global competition without headache. The MCP allows you to track keyword rankings across multiple regional markets using dedicated tools like scrape_google_serp and scrape_yandex_serp.
- Handle modern websites easily. If a site requires JavaScript to load content, the universal scraper tool handles the headless browser rendering for you, ensuring your data isn't incomplete.
- Consolidate multiple data types into one workflow. You can scrape not just products (scrape_google_shopping) but also images and news articles from the same platform in a single session.
- Gain control over advanced scraping parameters. The scrape_custom_payload tool lets you fine-tune proxy locations and configurations, overcoming common rate limits or geo-blocking issues.

## How It Works

The bottom line is you talk to your AI client like you're asking a teammate for data; it does the heavy lifting of crawling and formatting the results.

1. Subscribe to the Oxylabs MCP on Vinkius, then input your specific Oxylabs username and API credentials.
2. Instruct your AI client (Claude, Cursor, etc.) to perform a data extraction task, specifying the target website or search query.
3. The MCP executes the necessary scraping tool, returning structured JSON data containing the requested text, images, and metadata.

## Frequently Asked Questions

**How does Oxylabs MCP handle modern websites with JavaScript?**
The universal scraper tool handles this by simulating client-side execution. This means if the website loads content only after running code (a single-page application), the tool renders that code and captures the final HTML.

**Can Oxylabs MCP scrape international search engines?**
Yes, it supports multiple regions. You can use tools like scrape_yandex_serp to analyze rankings specifically for the CIS/RU market, complementing Google's global data.

**Does Oxylabs MCP only work on product pages?**
No. While it excels at e-commerce (scrape_amazon_product), you can also use scrape_google_news for time-indexed articles or scrape_google_images for visual assets.

**What is the difference between scrape_google_serp and scrape_universal?**
The Google Search tool provides structured, pre-parsed data specifically for SERP components (ads, snippets). The universal scraper gives you raw HTML from any arbitrary URL, requiring more post-processing.

**Can Oxylabs MCP help with proxy management?**
Yes. You can use the scrape_custom_payload tool to send custom JSON payloads that configure advanced proxy types and location settings for large-scale operations.