chrome-server-mcp

Chrome MCP Server

Chrome browser control connects AI and MCP clients to the CodingBaby extension through chrome-server-mcp. Its tools cover page interaction, tab management, screenshots, and sequential batches for workflows such as filling forms or running searches.

Server facts

chrome-server-mcp

Project

License
Not specified

Tech

Runtime / language
Not specified
Transports
stdio
Remote available
No
Authentication
Not specified

Activity

GitHub stars
29
Last commit
Data as of

Tags

  • Browser Automation
  • chrome server
  • mcp

Overview

Chrome navigation, page interaction, and tab handling are exposed to MCP clients through chrome-server-mcp and the CodingBaby companion extension. The server communicates with clients over stdio and uses WebSocket to reach the extension, which must be installed and enabled to carry out browser actions. Clients can open a URL, click at coordinates, enter text into a focused element, send keyboard input, scroll, and adjust viewport dimensions. Tab tools let them inspect open tabs, create another tab, switch by index, or close a tab.

For workflows involving several interactions, the batch tool runs browser operations sequentially in a single request, including combinations used for form entry, searches, and navigation. Visual feedback includes capturing a selected page region and waiting for a specified interval before receiving a screenshot. The README also describes full-page screenshots returned as Base64 image data, access to saved screenshots, and saving page HTML to a temporary server-side file. On the initial browser-tool call, the server attempts to start its WebSocket service and waits for the extension connection.

chrome-server-mcp setup notes

  • chrome-server-mcp records stdio only; confirm a network transport upstream before planning remote access.
  • chrome-server-mcp has no recorded license; confirm reuse terms in its repository.

Related MCP servers