Web Page to RAG Chunks

Convert any public web page to clean Markdown, heading-aware RAG chunks and metadata.

Extracts the main content with Mozilla Readability, converts it to Markdown with absolute links, and returns heading-aware chunks with token estimates plus title, language, canonical, dates and a sha256 content hash for change detection. robots.txt is respected.

$0.001 per call in USDC on Base via x402, or pay by card with prepaid credits (1 credit) · 1 free call per caller · no signup ·

GET https://freetier-agent-api.pages.dev/api/page-rag

When to use it

Use page-rag to feed a web page into an LLM or a vector store. Use page-meta or page-check for SEO/meta tags and HTTP status only. page-markdown is the older, simpler converter; page-rag is recommended.

Parameters

Example output (real response, trimmed)

{
 "url": "https://en.wikipedia.org/wiki/HTTP_402",
 "finalUrl": "https://en.wikipedia.org/wiki/HTTP_402",
 "status": 200,
 "contentType": "text/html",
 "ok": true,
 "title": "List of HTTP status codes - Wikipedia",
 "description": null,
 "siteName": null,
 "language": "en",
 "canonical": "https://en.wikipedia.org/wiki/List_of_HTTP_status_codes#402",
 "publishedAt": null,
 "modifiedAt": null,
 "jsonLdTypes": [],
 "markdown": "# List of HTTP status codes - Wikipedia\n\nThis article lists standard and notable non-standard [HTTP response status codes](https://en.wikipedia.org/wiki/HTTP...",
 "wordCount": 6573,
 "tokenEstimate": 18376,
 "truncated": true,
 "truncatedReason": "page HTML larger than 350000 bytes after removing scripts/styles; only the first part was converted",
 "likelyNeedsJavaScript": false,
 "contentHash": "sha256:7ad19acc51489e84fb2f495ed08b7ac0dc6a9e6f8635caefcc8cef78238a4c52",
 "chunks": [
  {
   "index": 0,
   "headingPath": "List of HTTP status codes - Wikipedia",
   "text": "# List of HTTP status codes - Wikipedia\n\nThis article lists standard and notable non-standard [HTTP response status codes](https://en.wikipedia.org/wiki/HTTP...",
   "tokenEstimate": 666
  },
  {
   "index": 1,
   "headingPath": "List of HTTP status codes - Wikipedia > Standard codes > 1xx informational response",
   "text": "The server has received the request headers and the client should proceed to send the request body (in the case of a request for which a body needs to be sen...",
   "tokenEstimate": 460
  }
 ],
 "extractor": "readability",
 "fetchMs": 484,
 "fetchedAt": "2026-10-24T18:11:05.434Z",
 "sources": [
  {
   "name": "Live fetch of the page you name (plain HTTP, robots.txt respected); Mozilla Readability + Turndown (Apache-2.0 / MIT), linkedom (ISC)",
   "url": "https://github.com/mozilla/readability"
  }
 ]
}

Try it

Free call (no wallet needed)

curl -s "https://freetier-agent-api.pages.dev/api/page-rag?url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FHTTP_402"

Pay by card (no wallet): prepaid credits

Buy a credit pack on Gumroad (1,000 credits $3), then send your license key. The X-Credits-Remaining response header shows your balance; failed calls are not charged.

curl -s "https://freetier-agent-api.pages.dev/api/page-rag?url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FHTTP_402" \
  -H 'X-License-Key: YOUR-LICENSE-KEY'

JavaScript / TypeScript (@x402/fetch)

// npm i @x402/fetch @x402/evm viem
import { wrapFetchWithPaymentFromConfig } from "@x402/fetch";
import { ExactEvmScheme } from "@x402/evm";
import { privateKeyToAccount } from "viem/accounts";

const account = privateKeyToAccount(process.env.EVM_PRIVATE_KEY); // wallet with USDC on Base
const fetchWithPayment = wrapFetchWithPaymentFromConfig(fetch, {
  schemes: [{ network: "eip155:8453", client: new ExactEvmScheme(account) }],
});
const res = await fetchWithPayment("https://freetier-agent-api.pages.dev/api/page-rag?url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FHTTP_402");
console.log(await res.json());

Python (x402 + httpx)

# pip install "x402[httpx,evm]" eth-account
import asyncio, os
from eth_account import Account
from x402 import x402Client
from x402.http.clients import x402HttpxClient
from x402.mechanisms.evm import EthAccountSigner
from x402.mechanisms.evm.exact.register import register_exact_evm_client

async def main():
    client = x402Client()
    register_exact_evm_client(client, EthAccountSigner(Account.from_key(os.environ["EVM_PRIVATE_KEY"])))
    async with x402HttpxClient(client) as http:
        r = await http.get("https://freetier-agent-api.pages.dev/api/page-rag?url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FHTTP_402")
        print(r.json())

asyncio.run(main())

A bare request without parameters returns the HTTP 402 payment challenge (x402 v2 in the PAYMENT-REQUIRED header, v1 in the body). Invalid input gets a 400 before any payment.

Billing policy

Failed calls are never charged. Failed calls are never charged: the payment is verified first, the request runs, and the payment is settled only if it succeeded. Errors (4xx/5xx) are not settled, and invalid input is rejected before any payment or free call.

Uptime and latency

Independent monitoring of this endpoint: 402index service page (health, 30-day uptime, p50 latency).

Also available as

• MCP tool page_rag at https://freetier-agent-api.pages.dev/mcp (Streamable HTTP; in-band x402 payment via _meta["x402/payment"], or card credits via the X-License-Key header)
• OpenAPI 3.1 with the full output JSON Schema

Agent Extraction API by FreeTier Kits · Terms · llms.txt · /.well-known/x402 · MCP: https://freetier-agent-api.pages.dev/mcp · Contact: tec@mail.grokbot.com
This site and its texts were written with AI assistance and reviewed for accuracy; example outputs are real responses from the API.