macOS Screen View & Control MCP Server

@jhead

Basic information

About macOS Screen View & Control MCP Server

MCP Server to screenshot and control macOS windows (made for Cursor)

macOS Screen View & Control MCP Server config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "macos-screen-mcp": {
      "command": "python",
      "args": [
        "-m",
        "macos_screen_mcp"
      ]
    }
  }
}

macOS Screen View & Control MCP Server tools

Captures a screenshot of a specific window by its title or ID

Lists all visible windows

Finds a window by title or owner name

Sends a keyboard key press event to the active window

Types a sequence of text characters

Overview

What is macOS Screen View & Control MCP Server?

A Model Context Protocol server for macOS that enables LLMs to capture screenshots of specific windows by title or ID, list visible windows, find windows, and send keyboard input. It is designed for developers and AI agents needing visual or interactive access to macOS windows.

How to use macOS Screen View & Control MCP Server?

Install via pip (pip install git+ssh://git@github.com/jhead/macos-screen-mcp.git) and run with python -m macos_screen_mcp. Configure the server URL (e.g., http://localhost:8000/sse) in Claude or Cursor settings under mcpServers. Use the provided tools: capture_window_screenshot, list_windows, find_window, send_key, and type_text.

Key features of macOS Screen View & Control MCP Server

Capture screenshots of windows by title or ID
List all visible windows on macOS
Find windows by title or owner name
Send keyboard key presses with modifier support
Type sequences of text with configurable keystroke delay

Use cases of macOS Screen View & Control MCP Server

An AI assistant screenshotting a specific application window for analysis
Automating window discovery and interaction on macOS
Sending keyboard shortcuts (e.g., Command+C) to active windows
Typing text into input fields programmatically
Debugging window state by listing visible windows

Frequently asked questions

What tools does macOS Screen View & Control MCP Server provide?

It provides `capture_window_screenshot`, `list_windows`, `find_window`, `send_key`, and `type_text`.

How do I install and run the server?

Install via `pip install git+ssh://git@github.com/jhead/macos-screen-mcp.git` and run with `python -m macos_screen_mcp`. Debug using `npx @modelcontextprotocol/inspector python -m macos_screen_mcp`.

What keys are supported for `send_key`?

Supported keys include letters a‑z, numbers 0‑9, return, tab, space, delete, escape, arrow keys, and modifier keys (command, shift, control, option, and their right variants).

Does the server support authentication?

No. The configuration uses a local HTTP URL (`http://localhost:8000/sse`) without any authentication mechanism mentioned.

On which operating systems does it run?

The server is explicitly for macOS; it uses macOS‑specific window APIs (e.g., window titles, screenshots) and does not support other operating systems.

Comments

More Other MCP servers

Volley.sh

Volley

A good sequence diagram can solve almost any communication problem in engineering. Volley lets your agent write them as plain text, with the real request and response payloads on every arrow. It checks its own syntax wi

Health by Dr Brad

Dr Brad Stanfield

Health by Dr Brad works out a health plan from your own results: which screening is due at your age, and evidence-based suggestions, each with the reason and the published guideline or study behind it. Your record is one

Hindsight

vectorize-io

Hindsight: Agent Memory That Learns

50.4K

Klarix Intelligence Engine

Klarix

B2B competitive intelligence, live sales battlecards, 5-axis ICP prospect fit scoring, and tech stack teardowns over MCP and REST with cited public-web evidence.

Screenpipe

screenpipe

YC (S26) | Open Computer History | Continuously record your company computer work, map your workflows, help you find work worth automating, and power your agents' context

22.0K