Developer ToolsFreeactiveReviewed by MCPIndex

Chrome DevTools MCP

Expose Chrome runtime, DOM, network, console, and performance diagnostics to AI coding agents.

Looking for more MCP servers? Browse the full MCP tools directory or explore more tools in Developer Tools.

Quick overview

Chrome DevTools MCP operates below the application abstraction layer, where browser failures actually occur. It lets an agent inspect network requests, console errors, DOM state, performance traces, and rendering behavior instead of guessing why a page is empty or slow. The critical edge case is that DOM output is not equivalent to application state: hydration, XHR, GraphQL, service workers, and redirects can mutate the result after the initial document loads. Filter diagnostics before returning them because network payloads and console logs may contain cookies, authorization headers, tokens, or user data.

What this MCP server is best for

  • Chrome DevTools MCP is useful when you want to extend an AI assistant with real tools and live system access.
  • It helps move from chat-only answers to real actions such as reading data, managing systems, or retrieving current information.
  • It is best for developer and technical workflows where AI needs controlled access to external tools or services.

When to choose it

Choose Chrome DevTools MCP when you want an MCP server focused on Developer Tools and need tighter integration with your existing tools.

Good fit

chromedevtoolsbrowser-debuggingnetworkperformance

Chrome DevTools MCP Configuration

Use the following configuration as a starting point for Claude Desktop or any compatible MCP client, then replace placeholder credentials with your own values.

claude_desktop_config.json
{
    "mcpServers": {
      "chrome-devtools": {
        "command": "npx",
        "args": [
          "-y",
          "chrome-devtools-mcp@latest"
        ]
      }
    }
  }

How to set up Chrome DevTools MCP

These setup steps cover the typical installation flow for this MCP server.

  1. 1

    Install Node.js 22 or a version supported by the installed Chrome DevTools MCP release.

  2. 2

    Install Google Chrome Stable and verify that the browser can launch under the current user.

  3. 3

    Add the Chrome DevTools MCP configuration to the MCP client.

  4. 4

    Start with read-only inspection of console errors, network requests, DOM readiness, and performance.

  5. 5

    Filter request URLs, headers, response bodies, local storage, and traces before returning them to the model.

  6. 6

    Pro-Tip: Never expose a browser profile containing saved passwords, cookies, or active production sessions. Use a dedicated Chrome profile and a domain allowlist.

  7. 7

    Pin the package version for CI or repeatable debugging instead of resolving @latest on every startup.

Frequently asked questions

Common questions for Chrome DevTools MCP.

What is the silent failure mode in Chrome DevTools MCP?

The browser can report document.readyState as complete while the application is still waiting for hydration or a GraphQL response. Wait for a required selector or response, assert record counts, detect interstitial markers, and fail with EXTRACTION_CONTRACT_FAILED when the expected state is absent.

Why can Chrome DevTools MCP return sensitive data?

Network logs, request headers, console output, local storage, and DOM snapshots can contain cookies, authorization headers, PII, and tokens. Redact those fields server-side and use an isolated browser profile before exposing diagnostics to an AI client.

Chrome DevTools MCP vs Competitors

FeatureChrome DevTools MCPCompetitor
Browser runtime diagnosis Inspects console, DOM, network, rendering, and performance state. Screenshot-only testing
Deterministic end-to-end testing Primarily diagnostic and requires filtered artifacts. Playwright test runner

Related Guides