omniparser-autogui-mcp

What is this MCP

This is a Model Context Protocol (MCP) server that uses Microsoft's OmniParser to analyze screen content and perform automated GUI operations on Windows systems.

How to use this MCP

Install by cloning the repository with submodules, setting up the environment variables, and configuring the MCP server in your claude_desktop_config.json file. The server can be customized with various environment variables for window targeting, backend configuration, and communication methods.

What this MCP can be used for

This MCP enables automated GUI operations like searching in on-screen browsers or interacting with applications by analyzing screen content with OmniParser's OCR and computer vision capabilities. It's particularly useful for automation tasks where direct API access isn't available.