MCP SERVER · OFFICIAL PROJECTPull clean, structured data off any site — including JavaScript-heavy and geo-restricted pages — with CAPTCHAs and proxies handled for you.
No longer maintained. The GitHub repository for this server is archived — no fixes or security updates are expected.
LABSCO SUMMARY
The split is the point: the Web Scraper API tools are for when you know the URL and want raw or parsed output, and the AI Studio tools are for when you describe what you want and let a crawler or browser agent go find it. You can hold both sets of credentials and reach for whichever fits, or hold just one and use only that half.
Oxylabs' official server, exposing two toolsets you can use together or apart: four Web Scraper API tools for direct scraping, and six AI Studio tools for prompt-driven extraction, crawling and a browser agent. It runs from PyPI over stdio, or as Oxylabs' hosted server with no install.
At least one Oxylabs credential, and which one depends on the tools you want: the Web Scraper API tools use OXYLABS_USERNAME and OXYLABS_PASSWORD (a one-week free trial is available), and the AI Studio tools use OXYLABS_AI_STUDIO_API_KEY (1,000 free credits). Use the hosted server at https://mcp.oxylabs.io/mcp with those credentials in request headers, or run it locally from the oxylabs-mcp PyPI package over stdio, which also needs the uv package manager.
One command — uvx oxylabs-mcp
USE CASES
Point it at a URL that renders in JavaScript or blocks datacenter traffic and get clean Markdown back — the rendering, CAPTCHA solving and country-level proxying are handled before the content reaches you.
Scrape an Amazon search page into parsed JSON, then open the individual products with variant auto-selection for accurate pricing, using the domain and locale the listing lives in.
Describe what you are after and let it crawl from a starting URL, gathering the same data from every matching page, capped at the number of sources you allow.
Hand the browser agent a prompt for a site that hides data behind clicks, forms or scrolling, and get back the extracted result rather than a raw dump you still have to parse.
Run a search and return the Markdown of each result in one step, so a research question is answered without chaining a search and a scrape yourself.
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
ai_browser_agentRun the browser agent and return the data in the specified format.
This tool is useful if you need navigate around the website and do some actions. It allows navigating to any url, clicking on links, filling forms, scrolling, etc. Finally it returns the data in the specified format. Schema is required only if output_format is json, csv or toon. 'task_prompt' describes what browser agent should achieve
ai_crawlerTool useful for crawling a website from starting url and returning data in a specified format.
Schema is required only if output_format is json, csv or toon. 'render_javascript' is used to render javascript heavy websites. 'return_sources_limit' is used to limit the number of sources to return, for example if you expect results from single source, you can set it to 1.
ai_mapTool useful for mapping website's URLs.
ai_scraperScrape the contents of the web page and return the data in the specified format.
Schema is required only if output_format is json or csv. 'render_javascript' is used to render javascript heavy websites.
ai_searchSearch the web based on a provided query.
'return_content' is used to return markdown content for each search result. If 'return_content' is set to True, you don't need to use ai_scraper to get the content of the search results urls, because it is already included in the search results. if 'return_content' is set to True, prefer lower 'limit' to reduce payload size.
amazon_product_scraperScrape Amazon products.
Supports content parsing, different user agent types, domain, geolocation, locale parameters and different output formats. Supports Amazon specific parameters such as currency and getting more accurate pricing data with auto select variant.
amazon_search_scraperScrape Amazon search results.
Supports content parsing, different user agent types, pagination, domain, geolocation, locale parameters and different output formats. Supports Amazon specific parameters such as category id, merchant id, currency.
generate_schemaGenerate a json schema in openapi format.
google_search_scraperScrape Google Search results.
Supports content parsing, different user agent types, pagination, domain, geolocation, locale parameters and different output formats.
universal_scraperGet a content of any webpage.
Supports browser rendering, parsing of certain webpages and different output formats.
BEFORE YOU INSTALL
10 tools when we connected. Locally, the tools shown depend on credentials: username and password expose the four Web Scraper API tools, the AI Studio key the other six. Over streamable HTTP all ten are listed, but calls without credentials fail.
CONTEXT COST
Rated by the size of the 10 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
uvx oxylabs-mcphttps://mcp.oxylabs.io/mcpAdd this hosted endpoint to your MCP client.
claude mcp add --transport http oxylabs https://mcp.oxylabs.io/mcp \
--header "Authorization: Basic $(echo -n '<YOUR_OXYLABS_USERNAME>:<YOUR_OXYLABS_PASSWORD>' | base64)" \
--header "X-Oxylabs-AI-Studio-Api-Key: <YOUR_OXYLABS_AI_STUDIO_API_KEY>"{
"mcpServers": {
"oxylabs": {
"url": "https://mcp.oxylabs.io/mcp",
"headers": {
"Authorization": "Basic <YOUR_BASE64_USERNAME>",
"X-Oxylabs-AI-Studio-Api-Key": "<YOUR_OXYLABS_AI_STUDIO_API_KEY>"
}
}
}
}{
"mcpServers": {
"oxylabs": {
"command": "uvx",
"args": [
"oxylabs-mcp"
],
"env": {
"OXYLABS_USERNAME": "<YOUR_OXYLABS_USERNAME>",
"OXYLABS_PASSWORD": "<YOUR_OXYLABS_PASSWORD>",
"OXYLABS_AI_STUDIO_API_KEY": "<YOUR_OXYLABS_AI_STUDIO_API_KEY>"
}
}
}
}Clone the repository:
git clone https://github.com/oxylabs/oxylabs-mcpThen add this block, with <YOUR_ABSOLUTE_PATH_TO_FOLDER> set to the folder that holds the clone:
{
"mcpServers": {
"oxylabs": {
"command": "uv",
"args": [
"--directory",
"/<YOUR_ABSOLUTE_PATH_TO_FOLDER>/oxylabs-mcp",
"run",
"oxylabs-mcp"
],
"env": {
"OXYLABS_USERNAME": "<YOUR_OXYLABS_USERNAME>",
"OXYLABS_PASSWORD": "<YOUR_OXYLABS_PASSWORD>",
"OXYLABS_AI_STUDIO_API_KEY": "<YOUR_OXYLABS_AI_STUDIO_API_KEY>"
}
}
}
}MCP servers
manchittlab
AStheTECH
scrapezy
Decodo