paid-md

Endpoints (3 live)

• POST /research — Multi-page research brief with cited sources. Crawl one URL and follow its links, or pass a list of URLs, and get a synthesized answer to your question as JSON with a sources array, optionally shaped by your JSON Schema. Use it to compare several pages, gather a list across a site, summarise a topic from multiple sources, or answer a question that spans documents. Up to 8 pages per call, typically 99% fewer tokens than reading the pages yourself. (0.05 USDC on Base)
• POST /extract — Extract structured JSON from any web page or raw text using your own JSON Schema. Web scraping to structured data: give a URL (or pasted text, HTML or Markdown) plus a schema and get back only the fields you asked for: prices, names, dates, specs, contact details, tables, product data. Handles pages too long to read in full. Returns strict JSON that matches the schema, no prose. Pay per call in USDC via x402. (0.01 USDC on Base)
• GET /md — Read any public web page as clean, LLM-ready Markdown. URL to markdown web scraper: fetches the page, strips navigation, ads, scripts, cookie banners and boilerplate, keeps the main content with headings, lists, links and tables. Use it to read, quote or summarise an article, docs page, blog post, product page or GitHub issue from a URL. Typically 90-99% fewer tokens than raw HTML; token counts returned in headers. (0.002 USDC on Base)