Show HN: LLM agents to extract web data

Hey this looks great, Do you guys also supporting parsing relatively bigger JSONs like amazon product pages or wikipedia pages ? How much ballpark $ amount does it take typically to scrape them ?