Tavily icon

Tavily

Search the web, extract page content, crawl and map websites, and run in-depth research with Tavily

Actions6

Extract → URLs

AI-generated

Summary

Extracts page content from one or more URLs by sending them to Tavily’s Extract API. Add the URLs to process, then optionally configure extraction depth, content format, images, favicon, usage details, or a query for focused content chunks.

Inputs

  • URLs (required) — One or more URLs to extract. In standard node mode, add URLs as separate values. In tool mode, provide URLs separated by commas or new lines; the model can define them, with the field guidance allowing up to 20 URLs.
  • Options: Extract Depth — Choose basic or advanced extraction. Advanced can retrieve more content, including tables and embedded content, and may take longer. In tool mode, extraction depth can also be set to Let the Model Decide; advanced costs twice as many credits.
  • Options: Format — Return extracted page content as Markdown (default) or plain text.
  • Options: Query — Describe the information to extract from the URLs. Enables chunk retrieval; each source returns 1–5 chunks, each up to 500 characters (default 3).
  • Options: Include Images — Include a list of images extracted from the URLs (default false).
  • Options: Include Favicon — Include each result’s favicon URL (default false).
  • Options: Include Usage — Include credit usage information in the response (default false).

Output shape

A list of n8n items created from Tavily’s response, associated with the input item that triggered the request.

The operation sends a POST request to /extract. It errors if no URLs are provided. Response fields depend on the URLs and selected options.

Links

Discussion