EU-native web scraping for AI agents & developers
🍮 crawlbrulee turns a url into clean markdown, a screenshot, metadata, links or html. we take care of the hard parts for you: headless Chrome for javascript-heavy pages, rotating proxies, retries and cookie banners, so you don't have to run and maintain your own scraper.
everything runs on 🇪🇺 EU servers, and our data processing agreement is public.
to try it, create a free account. you get 750 credits every month, no card needed. a simple page costs 1 credit.
curl -X POST https://api.crawlbrulee.com/api/scrape \
-H "Authorization: Bearer $CRAWLBRULEE_API_KEY" \
-H "Content-Type: application/json" \
-d '{ "url": "https://example.com", "extract": { "markdown": true, "links": true } }'you get json back with the markdown and the links. the shape is the same whether you wait for it or run it as a background job with a webhook. the getting started guide covers the rest.
our client tools live in this org, and all of them are open source:
| repo | what it is | try it |
|---|---|---|
| crawlbrulee-js | typed js/ts sdk, no runtime dependencies | npm install @crawlbrulee/sdk |
| crawlbrulee-py | python sdk, sync and async | pip install crawlbrulee |
| crawlbrulee-mcp | mcp server, so your agent gets scrape and map as tools | claude mcp add crawlbrulee -- npx -y @crawlbrulee/mcp |
| crawlbrulee-cli | scrape and map from your terminal | npx crawlbrulee scrape https://example.com |
| crawlbrulee-skills | agent skills that teach your coding agent how to use crawlbrulee | npx skills add crawlbrulee/crawlbrulee-skills |
| crawlbrulee-n8n | a verified n8n node | search "crawlbrulee" in n8n |
prefer your own client? the OpenAPI spec is public, so you can generate one in any language.
everything is Apache-2.0, except the n8n node, which is MIT.
- you don't pay for failures. a failed scrape costs 0 credits. so does a page we already have in cache.
- your data stays in the EU. the fetch, the render, the cache and the result all stay on EU servers. you pick the proxy exit, and if you pick an EU one, nothing leaves.
- output that models can read. markdown with ads, popups and cookie banners removed, and tall screenshots cut into slices an image model can handle. to drop anything else, like a nav or a footer, pass your own css selectors.
for a bug in one of the tools, open an issue in its repo. for anything else, write to contact@crawlbrulee.com or find us on X.
