Home / MCP / Chrome MCP Server

Chrome MCP Server

Provides an MCP interface that lets AI assistants control your existing Chrome browser for automated browsing, content analysis, and semantic search.

typescript

9.5kstars

View on GitHub

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
    "mcpServers": {
        "chrome_mcp_server": {
            "url": "http://127.0.0.1:12306/mcp"
        }
    }
}

Chrome MCP Server turns your active Chrome browser into a private, AI-controlled automation engine. It exposes browser functionality to AI assistants for complex browsing tasks, content analysis, and semantic search while preserving your existing session, settings, and login state.

How to use

You connect an MCP client to the server to start automating your browser. The server runs locally in your browser environment, letting you perform tasks like navigating pages, taking screenshots, analyzing content, managing bookmarks, and monitoring network activity through your preferred AI or chatbot client.

Key usage patterns you can follow include: controlling tabs and windows, executing scripted interactions on web pages, extracting and analyzing page content, and performing automated browsing workflows that combine multiple tools for a seamless assistant experience.

Available tools

chrome_navigate

Navigate to URLs and control viewport to set up the browsing context for automation.

chrome_switch_tab

Switch the current active tab to a target tab index or identifier.

chrome_close_tabs

Close specific tabs or entire windows to manage the browsing session.

chrome_go_back_or_forward

Navigate backward or forward in the browser history.

chrome_inject_script

Inject JavaScript into the current page to extend functionality.

chrome_send_command_to_inject_script

Send commands to a previously injected script to perform actions.

chrome_screenshot

Capture screenshots with support for targeting elements, full-page captures, and custom dimensions.

chrome_network_capture_start/stop

Start or stop capturing network traffic using the webRequest API.

chrome_network_debugger_start/stop

Attach a debugger to capture response bodies for network requests.

chrome_network_request

Send custom HTTP requests from the browser context.

search_tabs_content

AI-powered semantic search across content in open browser tabs.

chrome_get_web_content

Extract HTML or textual content from loaded pages.

chrome_get_interactive_elements

Find clickable elements on the page for automated interactions.

chrome_console

Capture and retrieve console output from browser tabs.

chrome_click_element

Click elements using CSS selectors to drive interactions.

chrome_fill_or_select

Fill form fields and select options to complete interactions.

chrome_keyboard

Simulate keyboard input and shortcuts for automation.

chrome_history

Search browser history with optional time-based filters.

chrome_bookmark_search

Find bookmarks by keywords to retrieve or manage them.

chrome_bookmark_add

Add new bookmarks, with folder organization.

chrome_bookmark_delete

Delete bookmarks from the browser store.