macOS Screen View & Control MCP Server
by jhead
This is a Model Context Protocol server that provides window screenshot capabilities on macOS. It enables LLMs to capture screenshots of specific windows by title or ID.
Last updated: N/A
What is macOS Screen View & Control MCP Server?
The macOS Screen View & Control MCP Server is a tool that allows Large Language Models (LLMs) to interact with macOS windows. It provides functionalities to capture screenshots, list windows, find windows, send key presses, and type text into applications.
How to use macOS Screen View & Control MCP Server?
First, install the server using pip install git+ssh://[email protected]/jhead/macos-screen-mcp.git. Then, run it as a script using python -m macos_screen_mcp. Configure your LLM (e.g., Claude or Cursor) to use the server by adding it to the mcpServers configuration with the provided name, URL, description, and version.
Key features of macOS Screen View & Control MCP Server
Capture window screenshots by title or ID
List all visible windows
Find a window by title or owner name
Send keyboard key press events to the active window
Type a sequence of text characters
Use cases of macOS Screen View & Control MCP Server
Automating tasks that require interacting with macOS applications
Capturing screenshots for visual analysis by LLMs
Controlling applications programmatically through keyboard inputs
Testing user interfaces by simulating user interactions
FAQ from macOS Screen View & Control MCP Server
What formats are supported for screenshots?
What formats are supported for screenshots?
The capture_window_screenshot tool supports binary and base64 output formats.
Can I specify a delay between keystrokes when typing text?
Can I specify a delay between keystrokes when typing text?
Yes, the type_text tool has an optional delay parameter to control the delay between keystrokes in seconds.
What modifier keys are supported?
What modifier keys are supported?
The supported modifier keys are command, shift, control, and option (including right_shift, right_option, right_control).
How do I contribute to the project?
How do I contribute to the project?
You can contribute by submitting pull requests with new tools, enhancements, bug fixes, or improvements to the documentation.
Under what license is this software released?
Under what license is this software released?
macos-screen-mcp is licensed under the MIT License.