Crawl4AI Plus icon

Crawl4AI Plus

Crawl web pages, extract data, and ask questions using Crawl4AI

Ask Question

AI-generated

Summary

Ask a question about one or more web pages by crawling a specified URL and using a Large Language Model (LLM) to generate a concise, structured answer with supporting details and source quotes.

Inputs

  • URL (required) — The URL to ask a question about.
  • Question (required) — The natural language question to answer based on the page content.
  • Crawl Scope — How extensively to crawl from the starting URL: 'Single Page' to crawl just the given URL, 'Follow Links' to crawl the URL plus discovered links (depth 1), or 'Full Site' to crawl recursively (depth 3). Defaults to 'Single Page'.
  • Options — Additional crawl options including Cache Mode, URL patterns to exclude (only applies to multi-page crawls), maximum pages to crawl, and CSS selector or JavaScript expression to wait for before extracting content. Cache Mode controls reading/writing of cache during crawling.

Output shape

a list of objects, each containing a structured JSON answer for the corresponding input item.

The response includes the original question, primary URL, domain, combined answer, supporting details, source quotes, success status, number of pages scanned, and crawl metrics. If no data is found, the answer indicates insufficient information. Supports multiple items as separate requests, each producing a respective result. Requires LLM credentials configured in the node to function.

Examples

Example 1: Ask what services a company offers by providing its homepage URL and a question.

Set URL to the homepage, question to 'What services do they offer?', Crawl Scope to 'Single Page' or wider, and optionally adjust caching and max pages.

Links

Discussion