Turn any URL into clean Markdown, HTML, or structured JSON.
Headless browser rendering, automated bot-detection evasion, and residential proxy rotation are executed server-side on every request. You send a URL; you receive LLM-ready data.
JavaScript Rendering
SPAs built with Next.js, React, Vue, or Angular render fully before extraction. The engine detects client-side hydration automatically.
Stealth Evasion
Bot-detection bypass, TLS fingerprint normalization, and automatic residential proxy escalation are included without extra surcharges.
Document Parsing
Ingest PDFs, whitepapers, DOCX, and XML feeds directly into clean text and tables, without configuring external OCR engines.
How SchemaFlow Processes the Web
Send a URL
One GET or POST request with target URL and optional flags for image extraction or document parsing.
Headless Render
Chromium cluster handles dynamic JavaScript, infinite scroll, and stealth residential proxy escalation.
DOM Sanitization
Ads, cookie banners, tracking scripts, and redundant markup are stripped to preserve core content hierarchy.
LLM Ingestion
Clean GitHub Flavored Markdown is delivered in <300ms, ready to be piped straight into your embeddings or agent context.
import { SchemaFlow } from '@schemaflow/sdk';
const client = new SchemaFlow({
apiKey: process.env.SCHEMAFLOW_API_KEY
});
// Convert any dynamic, JS-rendered URL into LLM-optimized Markdown
const { markdown, metadata } = await client.web.scrapeMarkdown({
url: 'https://linear.app/features',
waitForSelector: '[data-content]',
includeImages: true,
bypassAntiBot: true // Automated stealth residential escalation
});
console.log('Token count:', metadata.tokenCount);
console.log(markdown);