PlayMCP Browser Automation Server
A comprehensive MCP (Model Context Protocol) server for browser automation using Playwright. This server provides powerful tools for web scraping, testing, and automation.
Features
Core Browser Controls
- openBrowser - Launch a new browser instance with optional headless mode
- navigate - Navigate to any URL
- click - Click elements using CSS selectors
- type - Type text into input fields
- moveMouse - Move mouse to specific coordinates
- scroll - Scroll the page by specified amounts
- screenshot - Take screenshots of the page, viewport, or specific elements
- closeBrowser - Close the browser instance
- getPageSource - Get the complete HTML source code
- getPageText - Get the text content (stripped of HTML)
- getPageTitle - Get the page title
- getPageUrl - Get the current URL
- getScripts - Extract all JavaScript code from the page
- getStylesheets - Extract all CSS stylesheets
- getMetaTags - Get all meta tags with their attributes
- getLinks - Get all links with href, text, and title
- getImages - Get all images with src, alt, and dimensions
- getForms - Get all forms with their fields and attributes
- getElementContent - Get HTML and text content of specific elements
Advanced Capabilities
- executeJavaScript - Execute arbitrary JavaScript code on the page and return results
| Tool | Description | Required Parameters |
|---|
openBrowser | Launch browser instance | headless?: boolean, debug?: boolean |
navigate | Navigate to URL | url: string |
click | Click element | selector: string |
type | Type text into element | selector: string, text: string |
moveMouse | Move mouse to coordinates | x: number, y: number |
scroll | Scroll page | x: number, y: number |
screenshot | Take screenshot | path: string, type?: string, selector?: string |
getPageSource | Get HTML source | None |
getPageText | Get text content | None |
getPageTitle | Get page title | None |
getPageUrl | Get current URL | None |
getScripts | Get JavaScript code | None |
getStylesheets | Get CSS stylesheets | None |
getMetaTags | Get meta tags | None |
getLinks | Get all links | None |
getImages | Get all images | None |
getForms | Get all forms | None |
getElementContent | Get element content | selector: string |
executeJavaScript | Run JavaScript | script: string |
closeBrowser | Close browser | None |
Installation
Complete Installation Steps
- Prerequisites
- Node.js 16+ (download from nodejs.org)
- Git (for cloning the repository)
- Clone and Setup
git clone <repository-url>
cd PlayMCP
npm install
npm run build
- Install Playwright BrowsersThis downloads the necessary browser binaries (Chromium, Firefox, Safari).
- Verify InstallationYou should see "Browser Automation MCP Server starting..." if everything is working.
Quick Installation
git clone <repository-url>
cd PlayMCP
npm install && npm run build && npx playwright install
Usage
As MCP Server
Add to your MCP configuration file:
Standard MCP Configuration:
{
"servers": {
"playmcp-browser": {
"type": "stdio",
"command": "node",
"args": ["./dist/server.js"],
"cwd": "/path/to/PlayMCP",
"description": "Browser automation server using Playwright"
}
}
}
Alternative Configuration (works with VS Code GitHub Copilot):
{
"servers": {
"playmcp-browser": {
"type": "stdio",
"command": "node",
"args": ["/absolute/path/to/PlayMCP/dist/server.js"]
}
}
}
For Windows users:
{
"servers": {
"playmcp-browser": {
"type": "stdio",
"command": "node",
"args": ["C:\\path\\to\\PlayMCP\\dist\\server.js"]
}
}
}
VS Code GitHub Copilot Integration
This MCP server is fully compatible with VS Code GitHub Copilot. After adding the configuration above to your MCP settings, you can use all browser automation tools directly within VS Code.
Configuration Examples
Claude Desktop (config.json location):
- Windows:
%APPDATA%\Claude\config.json - macOS:
~/Library/Application Support/Claude/config.json - Linux:
~/.config/Claude/config.json
VS Code MCP Extension:
Add to your VS Code settings.json or MCP configuration file.
Example Full Configuration:
{
"mcpServers": {
"playmcp-browser": {
"type": "stdio",
"command": "node",
"args": ["/Users/username/PlayMCP/dist/server.js"],
"description": "Browser automation with Playwright"
}
}
}
Basic Web Scraping:
// Open browser and navigate
await openBrowser({ headless: false, debug: true })
await navigate({ url: "https://example.com" })
// Extract content
const title = await getPageTitle()
const links = await getLinks()
const forms = await getForms()
Form Automation:
// Fill out a form
await click({ selector: "#login-button" })
await type({ selector: "#username", text: "user@example.com" })
await type({ selector: "#password", text: "password123" })
await click({ selector: "#submit" })
Page Interaction:
// Scroll and interact
await scroll({ x: 0, y: 500 })
await moveMouse({ x: 100, y: 200 })
await click({ selector: ".dropdown-menu" })
Advanced JavaScript Execution:
// Run custom JavaScript
await executeJavaScript({
script: "document.querySelectorAll('h1').length"
})
// Modify page content
await executeJavaScript({
script: "document.body.style.backgroundColor = 'lightblue'"
})
// Extract complex data
await executeJavaScript({
script: `
Array.from(document.querySelectorAll('article')).map(article => ({
title: article.querySelector('h2')?.textContent,
summary: article.querySelector('p')?.textContent
}))
`
})
Screenshot and Documentation:
// Take screenshots
await screenshot({ path: "./full-page.png", type: "page" })
await screenshot({ path: "./element.png", type: "element", selector: "#main-content" })
Quick Start
- Install and setup:
git clone <repo-url> && cd PlayMCP
npm install && npm run build && npx playwright install
- Add to your MCP client configuration
- Start automating:
await openBrowser({ debug: true })
await navigate({ url: "https://news.ycombinator.com" })
const links = await getLinks()
console.log(`Found ${links.length} links`)
Development
- src/server.ts - Main MCP server implementation
- src/controllers/playwright.ts - Playwright browser controller
- src/mcp/ - MCP protocol implementation
- src/types/ - TypeScript type definitions
Requirements
System Requirements
- Node.js 16+ (LTS version recommended)
- Operating System: Windows, macOS, or Linux
- Memory: At least 2GB RAM (4GB+ recommended for heavy usage)
- Disk Space: ~500MB for browser binaries and dependencies
Dependencies
- Playwright: Handles browser automation (automatically installed)
- TypeScript: For compilation (dev dependency)
- Browser Binaries: Downloaded via
npx playwright install
Troubleshooting
Common Issues
- "Browser not initialized" error
- Make sure to call
openBrowser before other browser operations - Check if Node.js version is 16 or higher
- Playwright installation fails
# Try manual browser installation
npx playwright install chromium
# Or install all browsers
npx playwright install
- Permission errors on Linux/macOS
# Make sure the script is executable
chmod +x dist/server.js
- Path issues in MCP configuration
- Use absolute paths in the configuration
- On Windows, use double backslashes:
C:\\path\\to\\PlayMCP\\dist\\server.js - Verify the path exists:
node /path/to/PlayMCP/dist/server.js
- Browser crashes or timeouts
- Try running with
headless: false for debugging - Increase system memory if running multiple browser instances
- Check if antivirus software is blocking browser processes
Testing Your Installation
# Test the server directly
echo '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}' | node ./dist/server.js
You should see a JSON response listing all available tools.
License
MIT License