zlurp is a web scraping API built natively for AI agents. It converts any public URL into clean, structured markdown using the x402 payment protocol for per-request micropayments on Base mainnet.
The core idea: AI agents need to read web pages constantly, but every existing scraping tool requires accounts, API keys, and monthly subscriptions — none of which an autonomous agent can set up. zlurp uses x402 so any agent with a funded Base wallet can start scraping immediately with zero human setup.
zlurp is a sister product to docpull.ai, a PDF extraction API for AI agents built on the same x402 architecture.
Send a URL to /scrape. The server returns a 402 with USDC payment requirements. Your x402-compatible client pays $0.005 on Base, retries the request, and receives clean markdown. The entire flow takes one round trip with no human intervention.
zlurp uses Mozilla Readability for article extraction (the same engine as Firefox Reader View), Cheerio for full-page scraping, and Turndown for HTML-to-markdown conversion. Results are cached in Redis for one hour.
Email: [email protected]