**Give your agent a real browser.** Most "fetch a URL" tools hand your agent an empty page. Modern sites render with JavaScript, so a plain fetch gets nothing useful back. **`page_extract`** runs the page in a real headless Chrome, waits for the JavaScript, and returns clean text, headings, and every outbound link. **`site_extract`** does a whole documentation site — up to 25 pages of one host — in a **single call**. Same host only, and robots.txt is obeyed: anything it disallows is skipped and named in the result. **No API key. No account. No signup.** - **Free:** title, description, headings and the opening text — enough to tell whether it's the page you wanted. - **Paid:** $0.01 per page, $1.00 for a 25-page crawl. x402 per call in USDC on Arc, Base or Solana, or XNT on X1. Pass the signed payment straight to the tool — you never leave MCP to buy. Also included: a free token faucet your agent can claim with a signature (no gas, no captcha), exit-liquidity checks across Solana, Base and 70+ EVM chains, a live DEX screener, and a 3D city it can build in. Built by one person and one machine, in the open. Every number we publish is read from the chain, and our mistakes are published at apexfaucet.xyz/corrections.
First seen 2 Oct 2026. One server, whatever directories list it: each directory listing keeps its own page and history.
Tools
No tool list captured yet.
| Directory | Listing | Tier | First seen |
|---|---|---|---|
| Smithery | APEX Browser | - | 2 Oct 2026 |