mcp-playwright-scraper

A lightweight Node.js web scraper that uses Playwright to render JavaScript-heavy pages and converts them to clean Markdown.

Features

• 🎭 Playwright-powered headless browser — handles JS-rendered pages
• 📄 Readability for main content extraction
• ✨ Clean Markdown output via Turndown
• 🔓 Basic anti-scraping bypass (realistic user agent, viewport)
• 🚀 Zero config, single script

Requirements

• Node.js 18+
• Playwright Chromium browser

Installation

npm install
npx playwright install chromium

Usage

# Extract main article content (default)
node scrape.mjs https://example.com

# Full page content
node scrape.mjs https://example.com --full

Output

Outputs clean Markdown to stdout. Redirect to a file:

node scrape.mjs https://example.com > output.md

How it works

1. Launches a headless Chromium browser via Playwright
2. Navigates to the URL and waits for JS to render
3. Extracts the main content using Mozilla's Readability
4. Converts HTML to Markdown using Turndown
5. Outputs to stdout

License

MIT