An MCP server that exposes Chrome DevTools capabilities to AI coding assistants, enabling browser automation, network inspection, performance tracing, and debugging. It runs locally via Node.js and is designed for developers using MCP-compatible clients.
Project overview
It bridges AI coding agents and live Chrome DevTools, allowing automated actions to wait for results and capturing screenshots, traces, and source-mapped stack traces.
Project type
MCP · Workflow
Use cases
Coding & Development · Automation
Deployment
Refer to project documentation
License
Apache-2.0
Best for
Developers using AI coding assistants who need to automate, inspect, and debug live Google Chrome instances.
Key capabilities
Records Chrome DevTools traces and extracts actionable performance insights.
Analyzes network requests, takes screenshots, and checks browser console messages with source-mapped stack traces.
Uses Puppeteer to automate Chrome actions and automatically waits for action results.
Limitations and risks
Officially supports Google Chrome and Chrome for Testing only; operation in other Chromium-based browsers is not guaranteed.
Exposes the content of the browser instance to MCP clients; users should avoid sharing sensitive or personal information during sessions.
Getting started
Install Node.js LTS and Chrome, add configuration to an MCP client, and run npx -y chrome-devtools-mcp@latest.
Evidence and sources
README: Chrome DevTools for agents (`chrome-devtools-mcp`) lets your coding agent (such as Antigravity, Claude, Cursor or Copilot) control and inspect a live Chrome browser. It acts as a…
Release: chrome-devtools-mcp: v1.6.0
README: `chrome-devtools-mcp` exposes content of the browser instance to the MCP clients allowing them to inspect, debug, and modify any data in the browser or DevTools. Avoid sharing sen…
README: - **Get performance insights**: Uses [Chrome DevTools](https://github.com/ChromeDevTools/devtools-frontend) to record traces and extract actionable performance insights. - **Advan…