home / mcp / puppeteer debugger mcp server
Provides browser debugging, performance analysis, and memory inspection via a persistent Puppeteer‑based MCP server.
Configuration
View docs{
"mcpServers": {
"aliex7664-puppeteer-debugger-mcp-server": {
"command": "npx",
"args": [
"-y",
"@aliex7664/puppeteer-debugger-mcp@latest"
]
}
}
}You can run a Puppeteer‑based MCP Server to debug web pages, analyze performance, and inspect memory while keeping a persistent browser connection for faster, repeatable testing.
After you set up the MCP client to connect to the server, you can perform practical tasks such as navigating to pages, capturing console logs, inspecting element states, reviewing cached data, collecting performance metrics, profiling memory usage, and taking screenshots. All actions are invoked through MCP commands sent by your client, enabling automated debugging workflows and performance analysis within a single session.
Prerequisites you need before installing this MCP server are Node.js version 20 or higher and a Chrome or Chromium browser.
Install and start the MCP server from a configuration snippet that you load into your MCP client. Use the following exact command configuration to enable the server via npx. Then restart your MCP client to begin using the server.
The server is configured to run as a MCP server using a persistent browser session created at startup and kept running until the server stops. The configuration uses npx to fetch and run the latest version of the server package.
Config snippet to add to your MCP client configuration:
{
"mcpServers": {
"puppeteer_dbg": {
"command": "npx",
"args": ["-y", "@aliex7664/puppeteer-debugger-mcp@latest"]
}
}
}Navigate to a specified URL using the MCP client, enabling subsequent actions on that page.
Collect console errors, warnings, and log messages from the page, with optional URL and level filtering.
Check the state of a DOM element by selector, including its attributes, styles, visibility, and interactivity.
Retrieve LocalStorage, SessionStorage, Cookies, and IndexedDB status for a given page.
Fetch Performance Timeline data and page load metrics for performance analysis.
Capture a heap snapshot with optional parameters to control depth, garbage collection, and export modes.
Analyze current memory usage to identify potential leaks and memory utilization patterns.
Track object allocations over a set duration, with options to export the raw profile.
Take screenshots with options for full-page, output mode, and lazy-loading handling.
Generate a Lighthouse‑style performance report with configurable categories and audits.