Overview
What Chrome DevTools MCP does
Chrome DevTools MCP combines Puppeteer-driven page automation with debugging information from Chrome DevTools. An agent can reproduce an interaction, inspect console messages and requests, capture screenshots, and analyze performance traces in the same session. It is particularly strong when the objective is diagnosing a live web application rather than only completing a scripted page flow.
Best for
- Debugging frontend failures with console and network evidence
- Capturing and interpreting Chrome performance traces
- Inspecting a running Chrome instance from a coding agent
Not ideal for
- Cross-browser validation outside Chrome
- Sessions containing personal or confidential browsing data
- Basic automation where the smaller slim toolset or a test script is sufficient
Capabilities
What an agent can do
- 01
Chrome page automation
- 02
Console and network inspection
- 03
Screenshots and DOM snapshots
- 04
Performance trace recording and insights
- 05
CPU and network emulation
- 06
Connection to an existing debuggable Chrome instance
Representative tools and operations
list_pagesnavigate_pagetake_snapshotclickfilllist_console_messageslist_network_requestsperformance_start_trace
Installation
Connect Chrome DevTools MCP
Use the publisher’s current instructions as the source of truth. The examples below were checked on .
Generic MCP clientLaunch with npx
npx -y chrome-devtools-mcp@latest
{"command":"npx","args":["-y","chrome-devtools-mcp@latest"]}
VS CodeRegister from the command line
code --add-mcp '{"name":"io.github.ChromeDevTools/chrome-devtools-mcp","command":"npx","args":["-y","chrome-devtools-mcp"],"env":{}}'
Use --slim --headless when only the smaller automation surface is needed.
Trust and access
Authentication and security notes
Authentication: No separate service login is required. The server receives whatever access is present in the Chrome instance it launches or connects to; remote debugging endpoints may be configured with WebSocket headers.
Review before connecting
- The agent can inspect and modify content available to the connected Chrome profile.
- Use a dedicated browser profile without unrelated sessions or personal data.
- Usage statistics are enabled by default unless disabled by a flag or environment variable.
- Performance analysis may request Chrome UX Report field data unless the CrUX option is disabled.
Known limitations
- Official support targets Google Chrome and Chrome for Testing, not every Chromium derivative.
- It requires a compatible local Chrome and Node.js environment.
- Very large or complex browser sessions can increase startup and inspection overhead.
Evidence
Sources used for this guide
Facts were checked against primary publisher material. Descriptions and guidance are original Graphify summaries.
FAQ
Questions about Chrome DevTools MCP
How is Chrome DevTools MCP different from Playwright MCP?
Both can automate a browser, but Chrome DevTools MCP concentrates on Chrome-specific debugging and performance evidence, while Playwright MCP offers a broader automation model and multiple browser engines.
Can telemetry be disabled?
Yes. Start the server with --no-usage-statistics or set the documented environment variable; CrUX lookups have a separate opt-out flag.