capabilities.work

Extract Structured Data

A 3-step recipe with Hyperbrowser. We can help you get access.

Recipe · 3 steps

Extract Structured Data

with Hyperbrowser

Extract structured JSON from pages using a prompt and optional schema.

  1. Submit an extract job over the supplied URLs using the prompt and optional output schema, capturing the returned jobId.
  2. Poll the extract job status. The status moves through pending and running before reaching completed, failed, or stopped; loop back while still in progress and branch out on a terminal state.
  3. Fetch the completed extract job to return the structured data object the LLM produced from the pages.
What you'll need to have ready
  • apiKey: Hyperbrowser account API key sent in the x-api-key header.
  • urls: The list of URLs to extract structured data from.
  • prompt: The natural-language extraction prompt.
  • schema: Optional JSON schema describing the structured output shape.
For your developer

Arazzo workflow hyperbrowser:extract-structured-data, 3 operations: post-api-extract → get-api-extract-id-status → get-api-extract-id. Specs on APIs.io.

More recipes with Hyperbrowser

Browser Use Task RunCrawl Site and RetrieveHyperAgent Task RunPersistent Profile SessionSession LifecycleSession Recording RetrievalWeb Crawl and RetrieveWeb Search then Fetch