Rendex
Use Case

Turn Any URL into Clean Markdown for Your LLM

POST a URL, get back reader-mode Markdown, JSON, or article HTML — extracted from the fully-rendered page, so JavaScript-heavy sites work too. The fastest way to feed an AI agent or RAG pipeline clean text instead of raw markup.

The Problem

Feeding a model raw HTML wastes tokens on nav bars, scripts, ad slots, and cookie walls — and a plain fetch returns an empty shell for any site that renders client-side. Building your own extractor means running a headless browser, waiting for the SPA to hydrate, then bolting on a reader-mode library and maintaining it as sites change.

The Solution

Rendex renders the page first, then extracts the readable content for you. POST a URL to /v1/extract and choose Markdown, JSON, or article HTML; because extraction runs against the live DOM, single-page apps and lazy-loaded content come back as clean text. Need the screenshot and the text together? Set extract: true on /v1/screenshot/json and get image plus extracted content in a single call.

Get Started in Minutes

url-to-markdown.sh
curl -X POST https://api.rendex.dev/v1/extract \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://example.com/article",
    "extractFormat": "markdown",
    "blockCookieBanners": true
  }'
# => { "data": { "content": "# Title\n\n...", "title": "Title" } }
url_to_markdown.py
from rendex import Rendex

rendex = Rendex("YOUR_API_KEY")

# Render the page, then extract clean Markdown for the model.
doc = rendex.extract(
    url="https://example.com/article",
    extract_format="markdown",
    block_cookie_banners=True,
)

prompt = f"Summarize this article:\n\n{doc['content']}"
url-to-markdown.js
import Rendex from "@copperline/rendex";

const rendex = new Rendex("YOUR_API_KEY");

// Pull readable content from a JS-heavy page for a RAG index.
const { content, title } = await rendex.extract({
  url: "https://example.com/docs/guide",
  extractFormat: "markdown",
});

await indexDocument({ title, body: content });

100 free API calls/month — no credit card required. Get your API key and start building.

Built for This Use Case

Markdown, JSON, or HTML

Pick extractFormat per request — clean GitHub-flavored Markdown for LLMs, structured JSON for pipelines, or reader-mode article HTML.

Works on JS-Heavy Sites

Extraction runs against the fully-rendered DOM, so single-page apps, lazy-loaded sections, and client-rendered content come back as real text — not an empty shell.

Built for AI Agents & RAG

Hand a model clean Markdown instead of raw HTML — fewer wasted tokens, better grounding. Ideal for retrieval pipelines, summarizers, and research agents.

Image + Text in One Call

Set extract: true on /v1/screenshot/json to get the screenshot and the extracted content back together — capture and read the page in a single request.

Clean Reader Mode

Nav bars, scripts, cookie walls, and ad slots are stripped out. blockCookieBanners and hideSelectors remove the noise before extraction.

Edge Rendering

Rendered and extracted on Cloudflare's 300+ edge locations in seconds, with zero cold starts and no browser infrastructure to maintain.

Start Free, Scale as You Grow

100 calls/month free. Starter at $69/mo for 10K calls (or $58/mo billed annually — 2 months free). Pro at $179/mo for 100K calls.

Frequently Asked Questions

Related Resources