Book a call Get started

Endpoints

ScrapeClean markdown from one URL MapEvery URL on a site, one call CrawlWhole sites as one job SearchRanked results, already fetched ExtractTyped records from a schema BrandA company from its domain

Concepts

FormatsMarkdown, structured, links QualityConfidence on every response CreditsOne a page, failures free CachingFresh or cached, your call

Try it Run any endpoint on a URL of your own, no key

Web data for AI

Ground RAG in fresh contentCrawl a site on a schedule and pipe clean Markdown into your embeddings.Give an agent the live webSearch mid-answer and get results already fetched, cleaned and citable.Turn listings into a datasetPoint a JSON schema at a directory and get typed rows back.Watch pages for changesRe-run a set on a schedule and diff what came back.Track prices and competitorsMap a catalogue, pull every page, and extract the fields that move.

Company and brand data

Enrich a company from its domainOne call returns the name, industry, assets, palette and socials.Theme your app per tenantResolve a domain to a full design system and restyle your UI at runtime.Autofill onboardingTurn a work email domain into a filled-in company form.

All use cases Each one worked end to end, on real captures

DocsEndpoints, formats and limits BlogHow the engine is measured Extraction benchmarkWord-level F1 against the field Token indexWhat a page costs once it is clean

Talk to a human Thirty minutes, and bring your hardest URL

Reports

How the web behaves for machines

We read a lot of the web. These are the things we found that seemed worth measuring properly and publishing, with the method and the harness attached so you can check them.

ReproducibleMethod publishedOpen data

August 25th 2026

What websites cost an AI to read

Tokens of markup sent for every token of content delivered, measured across 40 well-known sites.

Read the report

918:1

heaviest measured

August 23rd 2026

Every extractor is good at articles

Word-level F1 for seven content extractors across 1,497 labelled pages. On articles they agree, and on everything else they do not.

Read the report

4x

wider off articles

[ Start ]

Clean web data is one call away.

500 free credits, no card required. Failures are never billed.

Get API key Read the docs

Success rate

Median scrape

ms