{"machine":"extract","slot":"A4","title":"Structured extraction (CSS selectors)","how_to_call":"POST https://innernetcorp.com/machines/extract with Content-Type: application/json","price_usd":"$0.003","currencies":["USDC","EURC"],"payment":"x402: send the POST without payment to receive a 402 with signed requirements; pay any option and retry.","input_example":{"url":"https://example.com","selectors":{"title":"h1","links":{"selector":"a","attr":"href","all":true}}},"input_schema":{"type":"object","required":["selectors"],"properties":{"url":{"type":"string","description":"Public http(s) URL to fetch. Provide url OR html."},"html":{"type":"string","description":"Raw HTML to extract from instead of fetching."},"base_url":{"type":"string","description":"Base URL for resolving when html is provided."},"selectors":{"type":"object","description":"Named selectors: value is a CSS string, or { \"selector\": \"...\", \"attr\": \"text|html|<attribute>\", \"all\": true }."}},"additionalProperties":false},"summary":"Scrape structured data from any public page: give a URL (or raw HTML) plus named CSS selectors, get clean JSON back. Supports text, attributes (href, src, ...), innerHTML, and all-matches arrays. The fast lane for agents that need facts off a page, not prose.","catalog":"/catalog"}