scrapeunblocker-mcp — Claude Code MCP server: setup and tools
Connects Claude Code to scraping tools that bypass Cloudflare and DataDome. It fetches fully rendered HTML, extracts clean Markdown, or returns structured JSON.
claude mcp add scrapeunblocker-mcp -e SCRAPEUNBLOCKER_KEY=<SCRAPEUNBLOCKER_KEY> -- npx -y scrapeunblocker-mcpWhat scrapeunblocker-mcp does
The server routes requests through ScrapeUnblocker to get past anti-bot screens like DataDome, Cloudflare, or Akamai. Its fetch_html tool grabs complete rendered pages, while fetch_markdown strips scripts and navigation. That leaves main content with absolute links at roughly 2-25% of the original HTML footprint. Context stays compact.
When sites require interaction, browse_page runs browser steps like clicking tabs, entering text, or scrolling to load content. These actions alter page state directly on the target site, so the tool is not read-only. If any step fails, the server reports the failure point along with the DOM state at that instant.
Over 100 specialized site plugins pull structured data directly from platforms such as Amazon, YouTube, or Booking.com. Without a valid key set in SCRAPEUNBLOCKER_KEY, all web fetching fails. Each execution counts as an API request against your account balance.
What it helps with
- extract page content as stripped Markdown to preserve model context
- bypass DataDome and Cloudflare bot barriers on target websites
- click buttons and fill input fields before capturing rendered HTML
- retrieve structured JSON from marketplaces and search engines
Example prompts for the agent
- Use
fetch_markdownto read the article at <url> and summarize its findings. - Run
list_elementson <url>, then usebrowse_pageto click the search button and grab the HTML. - Search Google for running shoes with
google_searchand display the organic results.
How to install scrapeunblocker-mcp
Platform CLI
claude mcp add scrapeunblocker-mcp -e SCRAPEUNBLOCKER_KEY=<SCRAPEUNBLOCKER_KEY> -- npx -y scrapeunblocker-mcpReplace <…> with your own values
Platform CLI
codex mcp add scrapeunblocker-mcp --env SCRAPEUNBLOCKER_KEY=<SCRAPEUNBLOCKER_KEY> -- npx -y scrapeunblocker-mcpReplace <…> with your own values
Config file
{
"mcpServers": {
"scrapeunblocker-mcp": {
"command": "npx",
"args": [
"-y",
"scrapeunblocker-mcp"
],
"env": {
"SCRAPEUNBLOCKER_KEY": "<SCRAPEUNBLOCKER_KEY>"
}
}
}
}Replace <…> with your own values
Config file
{
"servers": {
"scrapeunblocker-mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"scrapeunblocker-mcp"
],
"env": {
"SCRAPEUNBLOCKER_KEY": "<SCRAPEUNBLOCKER_KEY>"
}
}
}
}Replace <…> with your own values
Platform CLI
gemini mcp add -e SCRAPEUNBLOCKER_KEY=<SCRAPEUNBLOCKER_KEY> scrapeunblocker-mcp npx -y scrapeunblocker-mcpReplace <…> with your own values
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about scrapeunblocker-mcp
How do I install scrapeunblocker-mcp in Claude Code?
Run claude mcp add scrapeunblocker-mcp -e SCRAPEUNBLOCKER_KEY=<SCRAPEUNBLOCKER_KEY> -- npx -y scrapeunblocker-mcp. Replace the placeholder with your personal key from the ScrapeUnblocker dashboard.
What does scrapeunblocker-mcp do?
It gives your AI client tools to scrape websites through anti-bot systems. The server can return raw HTML, Markdown summaries, AI-parsed JSON, or interact with forms inside a browser.
What is required before using the tools?
You need an active ScrapeUnblocker account and an API key. Pass it via the SCRAPEUNBLOCKER_KEY environment variable during client configuration.
