WaterCrawl

Feeding web pages to a language model usually means writing a scraper and cleaning the output yourself.

WaterCrawl is a web crawling and content extraction platform that turns sites into structured data for AI applications, and it exposes everything through an API.

The free plan gives you 1,000 page credits, 100 per day, one seat, a crawl depth of 2, up to 50 pages per crawl and one concurrent crawl, with results kept for 7 days.

That is enough to test an idea, not to run a pipeline.

The Startup plan costs 4.80 a month billed yearly at 57.60, with 120,000 credits a year, 3 seats and depth 4.

Business is 79.99 a month and raises limits to 10 seats, depth 10, 2,500 pages per crawl and unlimited concurrent crawls, with priority support.