Engineering
chrome-devtools-mcp avatar

chrome-devtools-mcp

Chrome DevTools MCP server for AI-driven browser automation, testing, and debugging via Puppeteer. Features input automation, visual snapshots, performance tracing, and network inspection.

Introduction

The Chrome DevTools MCP server provides AI agents with deep, programmatic control over a live Chrome or Chromium instance using the Chrome DevTools Protocol (CDP) via Puppeteer. This skill is designed for software engineers, QA testers, and automation specialists who need to perform complex web interactions, verify UI components, or analyze browser-level performance metrics without manual intervention. By integrating directly into the OpenClaw environment, it enables agents to act as autonomous testers or scrapers, capable of navigating pages, handling authentication flows, and inspecting request/response cycles in real-time.

  • Full input automation including clicks, drag-and-drop, form filling, key presses, and file uploads.

  • Advanced navigation management to handle tabs, wait for specific network idle states, and control page selection.

  • Visual verification tools, including high-fidelity screenshots and DOM accessibility snapshots for layout debugging.

  • Performance auditing capabilities that record and analyze Chrome performance traces, including automated insight generation.

  • Comprehensive network inspection to monitor traffic, extract request details, and debug API interactions within the browser.

  • Device emulation for mobile testing, responsive design validation, and viewport resizing.

  • Console debugging access to read logs, stack traces, and evaluate arbitrary JavaScript in the page context.

  • Prerequisites include Node.js v20.19+ and a Chrome-based browser.

  • Best deployed in headless mode for server-side automation tasks or integrated with existing remote debugging ports.

  • Supports configuration for telemetry opt-out using --no-usage-statistics and --no-performance-crux flags.

  • Users should be aware that session data involves interacting with live web content and should avoid processing sensitive credentials in uncontrolled environments.

  • Ideal for browser testing, web automation, performance analysis, form filling, and visual regression testing workflows.

Repository Stats

Stars
4,454
Forks
1,214
Open Issues
7
Language
Python
Default Branch
main
Sync Status
Idle
Last Synced
Apr 30, 2026, 09:23 AM
View on GitHub