— Use case

Turn web pages into clean structured documents

Turn web pages into clean structured documents Crawl, extract, and deliver structured web intelligence via API.

Start free API docs

Pain points

  • PDF/HTML hybrids break parsers
  • Footers and nav pollute LLM context
  • Need blocks, not raw DOM

Architecture

  1. Fetch URL
  2. Strip boilerplate
  3. Return content[] blocks and metadata
  4. Optional export to Parquet

Example output

{ "title": "Annual report", "content": ["section 1..."], "links": [] }

FAQ

How fast can I start?

Sign up free, create an API key, and call /graph/domain-context or /scrape in minutes. See /docs for curl examples.

Is output AI-ready?

Yes — structured JSON, context_for_ai summaries, and link graphs designed for agents and RAG pipelines.