MCP: Crawlbase MCP
by Crawlbase
Official Crawlbase MCP server giving agents live web access — fetch any URL as raw HTML, clean Markdown or a screenshot, with JS rendering and proxy rotation handled for you.
Official Oxylabs MCP server: live web search, company lookup, and page scraping to Markdown or schema-defined JSON, including JavaScript-heavy and bot-protected sites, plus Oxylabs' dedicated scrapers.
Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.
Oxylabs Web API MCP is the official Model Context Protocol server for the Oxylabs Web API, Oxylabs' web data and scraping service. It gives an agent live web access through one connection: search the web, read any page, and call the dedicated scraper endpoints Oxylabs maintains, without the agent having to deal with proxies, rendering or blocking.
It has seven tools, all marked read-only so clients can run them without prompting. search returns ranked organic results. search_companies finds companies by name or description and returns profiles with industry, size, headquarters, funding, revenue and web traffic. scrape reads one URL or a list as Markdown by default, including JavaScript-heavy and bot-protected pages, and can return a screenshot or pull named fields as JSON from a prompt or a JSON schema. list_scrapers and run_scraper expose the API's dedicated scrape endpoints with their own parameters, check_requests collects background requests, and read_result pages through results too long for one reply. A bundled web-research skill is served as an MCP resource and a web_research prompt.
Oxylabs hosts the server at https://mcp.oxylabs.io/web-api-mcp (streamable HTTP), authenticated with a Web API key in an Authorization: Bearer header, so it needs a client that can send custom headers, such as Claude Code, Codex or Cursor. Clients limited to OAuth connectors, like claude.ai custom connectors, use the self-hosted server instead: the MIT-licensed Python package oxylabs-web-api-mcp (Python 3.10+, installed with uv or pipx), which runs over stdio or as a shared HTTP server and can cap per-key spend. Usage is billed by Oxylabs on the Web API plan behind the key.
by Crawlbase
Official Crawlbase MCP server giving agents live web access — fetch any URL as raw HTML, clean Markdown or a screenshot, with JS rendering and proxy rotation handled for you.
Official Browserless MCP server: scrape, search, map and crawl sites, run Lighthouse audits and custom Puppeteer code, and drive persistent cloud browser sessions from any MCP client.
Jina AI's official remote MCP server: read pages and PDFs as Markdown, search the web, arXiv and SSRN, take screenshots, rerank and deduplicate.
by SerpApi
Official SerpApi MCP server: real-time Google, Bing, YouTube and other search results as structured JSON or token-saving markdown.