Crawl a website: fetch a page, follow the links inside the same host and return every page it reached, within a page budget you set. For reading documentation, a product catalogue or a changelog in one paid call instead of many.