The official WebScraping.AI community node adds JavaScript rendering, rotating proxies, and AI-powered extraction to your n8n workflows — everything the built-in HTTP Request node can't do.
2,000 free API credits · No credit card required
n8n's built-in HTTP Request and HTML nodes are perfect for APIs and simple pages. Real-world websites fight back.
The HTTP Request node fetches raw HTML. On React/Vue/Angular sites and anything with lazy loading, the data you want isn't in the response at all.
Datacenter IPs — including n8n Cloud's shared ones — get challenged or blocked outright by Cloudflare, DataDome, and friends.
Loop over a few hundred URLs from one IP and rate limiting kicks in. Building proxy rotation inside n8n is a project of its own.
The WebScraping.AI node moves all of that behind one node: headless Chromium rendering, rotating datacenter/residential/stealth proxies, and retries — you only pay for successful requests.
Use the community node for the full experience, or plain HTTP Request if you can't install nodes.
Every operation supports JS rendering, proxy type & country, device emulation, custom headers, timeouts, and wait_for selectors.
On n8n Cloud plans where community nodes aren't available, point the built-in HTTP Request node at the API directly:
Method: GET
URL: https://api.webscraping.ai/html
Query parameters:
api_key = YOUR_API_KEY
url = https://example.com
js = true
proxy = residential
country = usSwap /html for /text (clean text), /ai/question (+ question param), or /ai/fields (+ fields) — same parameters everywhere. Full API reference.
Most n8n scraping guides end with an HTML node full of selectors that break on the next redesign. The AI Question and AI Extract Fields operations read the rendered page with an LLM instead — describe what you want in plain language and get clean JSON into the next node.
// AI Extract Fields operation
URL: https://competitor.com/product/123
Fields: {
"name": "Product name",
"price": "Current price with currency",
"stock": "In stock? true/false"
}
// Output (ready for the next node)
{
"name": "Acme Widget Pro",
"price": "$49.00",
"stock": "true"
}The four patterns behind most n8n scraping workflows
Check competitor prices on a schedule and alert the team when they move.
Pull page text, summarize or classify it with your LLM node, log the result.
Extract company and contact details from directory listings into your CRM.
Collect jobs, properties, or products from several sites into one database.
The fixes for the errors every n8n scraper hits
403 Forbidden or Cloudflare challenge page
The target is blocking datacenter IPs. Set Proxy Type to Residential (10–25 credits/request) in the node's proxy settings. Failed attempts aren't billed, so switching costs nothing.
The data is missing from the response
The page renders client-side. Make sure Enable JavaScript is on (it is by default), and set Wait For Selector to an element that appears once content loads — e.g. .price.
Timeouts on slow pages
Raise the node's Timeout (up to 30,000 ms) and the JavaScript Timeout for heavy single-page apps.
Only the first URL gets scraped
n8n passes items one at a time — put a Loop Over Items node before the WebScraping.AI node, or feed it a list and let it iterate item-by-item.
Hitting concurrency limits
Plans include a concurrency cap (2 on free accounts). Add a small Wait node in the loop or upgrade the plan for parallel workflows.
Does the WebScraping.AI node work on n8n Cloud?
Community node availability on n8n Cloud depends on your plan and n8n's verification rules. If you can't install community nodes, the built-in HTTP Request node works with the same API — set the URL to https://api.webscraping.ai/html with api_key, url, js, and proxy query parameters.
Can n8n scrape JavaScript-heavy websites with this integration?
Yes. Every operation can render pages in headless Chromium before returning content, with a configurable JS timeout and wait_for selector — this is exactly what n8n's built-in HTTP Request node can't do.
How do I avoid getting blocked while scraping in n8n?
Route requests through the node's proxy settings: rotating datacenter proxies for easy targets, residential (10–25 credits) for protected ones, and stealth (50 credits) for the hardest sites. IP rotation, retries, and browser fingerprinting are handled by the API, and failed requests are free.
What does scraping in n8n cost with WebScraping.AI?
Plans start at $29/mo for 250,000 credits; a plain request is 1 credit, JS rendering 5, residential proxies 10–25, and AI extraction adds 5. A free account includes 2,000 credits per month — enough to build and test workflows before paying anything.
Can I extract structured data without writing CSS selectors?
Yes — the AI Extract Fields operation takes plain-language field descriptions and returns JSON, and AI Question answers free-form questions about a page. Selector-based operations (Get Selected HTML) are also available when you want exact control.
Can I scrape pages behind a login with n8n?
Only in a limited way — you can pass session cookies via custom headers, but there's no interactive login flow. For fully interactive multi-step sessions you'd need a browser automation tool instead.
Get started with 2,000 free API credits. No credit card required.