About macOS Screen View & Control MCP Server
macOS Screen View & Control MCP Server exposes macOS window screenshots and keyboard input for LLMs that need to inspect or operate desktop windows. It can locate windows by title or ID, then send key presses or text.
Key features
- 01List visible windows
- 02Capture a window by title or ID
- 03Find windows by title or owner
- 04Send key presses to the active window
- 05Type sequences of text
Use cases
- 01Capture a specific macOS window
- 02Locate a target application window
- 03Send a keyboard shortcut to the active window
- 04Enter text into a window
MCP support
It is an MCP server that exposes window listing, search, and screenshot tools to MCP clients, along with keyboard input and text typing capabilities.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEmacOS Screen View & Control MCP Server
A Model Context Protocol server that provides window screenshot capabilities. This server enables LLMs to capture screenshots of specific windows on macOS, either by window title or window ID.
Available Tools
-
capture_window_screenshot- Captures a screenshot of a specific window by its title or IDwindow_identifier(string, required): Window title to search for or window IDsearch_in_owner(boolean, optional): Whether to search in window owner names (default: true)format(string, optional): Output format (binary or base64) (default: "binary")
-
list_windows- Lists all visible windows- No parameters required
-
find_window- Finds a window by title or owner nametitle(string, required): Window title or owner name to search forsearch_in_owner(boolean, optional): Whether to search in window owner names (default: true)
-
send_key- Sends a keyboard key press event to the active windowkey(string, required): The key to press (e.g., 'a', 'return', 'space')modifiers(list of strings, optional): List of modifier keys to hold (e.g., ['command', 'shift'])
-
type_text- Types a sequence of text characterstext(string, required): The text to typedelay(float, optional): Delay between keystrokes in seconds (default: 0.1)
Supported Keys
The following keys are supported:
- Letters: a-z (case-insensitive)
- Numbers: 0-9
- Special keys: return, tab, space, delete, escape
- Arrow keys: up_arrow, down_arrow, left_arrow, right_arrow
- Modifier keys: command, shift, control, option (also right_shift, right_option, right_control)
Examples
Send a single key press:
await send_key("return")
Send a key with modifiers:
await send_key("c", ["command"]) # Command+C (copy)
Type text:
await type_text("Hello, World!")
Installation
Using pip
Install macos_screen_mcp via pip:
pip install git+ssh://[email protected]/jhead/macos-screen-mcp.git
After installation, you can run it as a script using:
python -m macos_screen_mcp
Configuration
Configure
Add to your Claude or Cursor settings:
"mcpServers": {
"macos-screen": {
"name": "macos-screen",
"url": "http://localhost:8000/sse",
"description": "MCP server for capturing window screenshots",
"version": "1.0.0"
}
}
Debugging
You can use the MCP inspector to debug the server:
npx @modelcontextprotocol/inspector python -m macos_screen_mcp
Contributing
We encourage contributions to help expand and improve macos-screen-mcp. Whether you want to add new tools, enhance existing functionality, or improve documentation, your input is valuable.
Pull requests are welcome! Feel free to contribute new ideas, bug fixes, or enhancements to make macos-screen-mcp even more powerful and useful.
License
macos-screen-mcp is licensed under the MIT License. This means you are free to use, modify, and distribute the software, subject to the terms and conditions of the MIT License. For more details, please see the LICENSE file in the project repository.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.