Puppeteer MCP Server
Browser automation for web scraping, screenshots, and PDF generation
Puppeteer MCP Server gives AI agents the ability to control a headless Chrome browser for web scraping, taking screenshots, generating PDFs, clicking elements, filling forms, and navigating pages. It bridges the gap between AI reasoning and real-time web interaction, enabling tasks like automated testing and data extraction from dynamic websites.
Puppeteer MCP Server Configuration
{
"mcpServers": {
"puppeteer": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-puppeteer"]
}
}
}How to setup Puppeteer MCP Server?
- 1
Ensure Node.js 18+ is installed. Puppeteer will automatically download a compatible Chromium binary.
- 2
Add the configuration JSON above to your Claude Desktop config file.
- 3
Restart Claude Desktop to load the Puppeteer server.
- 4
Test by asking Claude to navigate to a website and take a screenshot.
- 5
For headful mode (visible browser), set the environment variable HEADLESS=false.
Is Puppeteer MCP Server free?
Yes, completely free and open-source under MIT license. Chromium is downloaded automatically at no extra cost.
How to setup Puppeteer MCP Server?
Simply add the npx configuration to your Claude Desktop config and restart. No API keys needed. Puppeteer downloads its own Chromium browser automatically on first run.
Can Puppeteer MCP Server handle JavaScript-rendered pages?
Yes, since it runs a real Chromium browser, it fully supports JavaScript execution, SPAs, and dynamically loaded content. This is its main advantage over simple HTTP-based scrapers.
Puppeteer MCP Server vs Competitors
| Feature | Puppeteer MCP Server | Competitor |
|---|---|---|
| JavaScript Rendering | ✅ Yes | ❌ No |
| Screenshot Capture | ✅ Yes | ❌ No |
| PDF Generation | ✅ Yes | ❌ No |
| Form Filling | ✅ Yes | ❌ Partial |
| Lightweight | ❌ No | ✅ Yes |
Tags
Are you the developer?
Claim this profile to update data, add documentation, view lead analytics, and control your tool's listing on MCPIndex.