> ## Documentation Index
> Fetch the complete documentation index at: https://docs.reader.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Quickstart

> Make your first scrape with self-hosted Reader in 60 seconds.

By the end of this page you'll have scraped a real page with the self-hosted Reader library and seen clean markdown come back.

## Install

```bash theme={null}
npm install @vakra-dev/reader
```

Make sure you're on Node 22.12.0+ and have Chrome's system dependencies if you're on Linux. See [Installation](/self-hosted/getting-started/installation) for details.

## Your first scrape

Create a file and paste this:

```javascript theme={null}
import { ReaderClient } from "@vakra-dev/reader";

const reader = new ReaderClient({ verbose: true });

try {
  const result = await reader.scrape({
    urls: ["https://example.com"],
    formats: ["markdown"],
  });

  const page = result.data[0];
  console.log(`Title:    ${page.metadata.website.title}`);
  console.log(`Engine:   ${page.metadata.engine}`);
  console.log(`Duration: ${page.metadata.duration}ms`);
  console.log();
  console.log(page.markdown);
} finally {
  await reader.close();
}
```

Run it:

```bash theme={null}
node your-file.js
```

You'll see Reader initialize, render the page in a headless browser, extract the main content, convert it to markdown, and print the result.

## What just happened

1. **`new ReaderClient(...)`** - creates a client. No browser yet.
2. **`reader.scrape(...)`** - on the first call, Reader spins up a browser pool behind the scenes and runs the scrape. The pool stays alive for subsequent calls.
3. **Playwright engine** - Reader renders the page in headless Chrome with JavaScript execution and proxy routing. Use `proxyTier: "premium"` to route through residential proxies for hostile sites.
4. **Content cleaning** - by default, Reader extracts only the main content, strips ads and navigation, and converts to clean markdown.
5. **`reader.close()`** - shuts down browsers. Optional - Reader also auto-cleans on `SIGTERM` / `SIGINT`.

## Try something more interesting

Scrape multiple URLs in parallel with progress tracking:

```javascript theme={null}
import { ReaderClient } from "@vakra-dev/reader";

const reader = new ReaderClient({
  browserPool: { size: 3 },
});

const result = await reader.scrape({
  urls: [
    "https://example.com",
    "https://example.org",
    "https://example.net",
  ],
  formats: ["markdown"],
  batchConcurrency: 2,
  onProgress: ({ completed, total, currentUrl }) => {
    console.log(`[${completed}/${total}] ${currentUrl}`);
  },
});

console.log(`
  Success: ${result.batchMetadata.successfulUrls}/${result.batchMetadata.totalUrls}
  Total:   ${result.batchMetadata.totalDuration}ms
`);

await reader.close();
```

## Where to go next

<CardGroup cols={2}>
  <Card title="Examples" icon="code" href="/self-hosted/getting-started/examples">
    Crawling, proxy rotation, dynamic content, and more.
  </Card>

  <Card title="Concepts: Scraping Engine" icon="lightbulb" href="/self-hosted/concepts/engine-waterfall">
    How the Playwright engine and proxy tiers work.
  </Card>

  <Card title="Guides: Batch Scraping" icon="layer-group" href="/self-hosted/guides/batch-scraping">
    Tune concurrency, handle errors, track progress.
  </Card>

  <Card title="API Reference" icon="book" href="/self-hosted/api-reference/reader-client">
    Full type reference for every option.
  </Card>
</CardGroup>