macOS Automation MCP Server
Enables macOS automation through 45 AppleScript-powered tools for system control, app management, calendar, reminders, notifications, clipboard, and task scheduling.
README
macOS Automation MCP Server
A comprehensive MCP (Model Context Protocol) server for macOS automation via AppleScript. This server provides 45 universal tools for system control, application management, calendar, reminders, notifications, clipboard operations, and scheduled tasks.
Built with FastMCP and Pydantic for type-safe, production-ready operation.
✅ Verified Working
- macOS 26.1+ compatible
- All 45 tools registered and functional
- AppleScript execution engine tested
- Shell command safety validation active
Features
P0 - Essential Tools (12 System + 6 App + 3 Notification + 3 Clipboard = 24 tools)
- System Control: Volume, dark mode, display sleep, WiFi, text-to-speech, screenshots
- Application Control: Launch, quit, list, and manage running applications
- Notifications: Send notifications, alerts, and dialogs
- Clipboard: Read, write, and manage clipboard contents
P1 - Extended Tools (6 Finder + 3 Calendar + 5 Reminders = 14 tools)
- Finder Operations: Create folders, move to trash, reveal files, search
- Calendar Integration: List, create, and manage calendar events
- Reminders Integration: List, create, complete, and manage reminders
P2 - Advanced Tools (4 Script + 3 Scheduler = 7 tools)
- Script Execution: Run AppleScript and shell commands
- Task Scheduler: Create and manage launchd scheduled tasks
Installation
Prerequisites
- macOS 10.15 or later
- Python 3.10 or later
- uv (recommended) or pip
Install with uv (Recommended)
cd /path/to/mcp-applescript-automation
uv sync
Install with pip
cd /path/to/mcp-applescript-automation
pip install -e .
Configuration
Claude Desktop / Cursor
Add to your MCP configuration file:
{
"mcpServers": {
"macos-automation": {
"command": "uv",
"args": ["run", "--directory", "/path/to/mcp-applescript-automation", "python", "main.py"],
"env": {}
}
}
}
Minerva Platform
Add to config/mcp-servers.json:
"macos-automation-mcp": {
"command": "uv",
"args": ["run", "--directory", "/Volumes/dev-4tb/AA-GitHub/MCP/mcp-applescript-automation", "python", "main.py"],
"env": {},
"description": "macOS Automation MCP: System control, app management, calendar, reminders, notifications, scheduled tasks via AppleScript"
}
Permissions
This server requires certain macOS permissions to function:
-
Automation Permission (Required)
- System Preferences > Privacy & Security > Privacy > Automation
- Grant permission to your terminal app (Terminal, iTerm2, Cursor, etc.)
-
Accessibility Permission (For some UI automation)
- System Preferences > Privacy & Security > Privacy > Accessibility
- Required for advanced UI interactions
-
Calendar/Reminders Access (For calendar/reminder tools)
- System Preferences > Privacy & Security > Privacy > Calendars/Reminders
- Required for calendar and reminder management
Available Tools (45 Total)
System Control (12 tools)
| Tool | Description |
|---|---|
get_system_info |
Get macOS version, hostname, user, and uptime |
set_volume |
Set system volume (0-100) |
get_volume |
Get current volume level |
toggle_mute |
Toggle audio mute |
toggle_dark_mode |
Toggle dark/light mode |
set_dark_mode |
Set dark mode explicitly |
is_dark_mode |
Check if dark mode is enabled |
sleep_display |
Put display to sleep |
say_text |
Speak text using TTS |
take_screenshot |
Capture screenshot |
get_wifi_network |
Get current WiFi network |
toggle_wifi |
Enable/disable WiFi |
Application Control (6 tools)
| Tool | Description |
|---|---|
list_running_apps |
List all running applications |
get_frontmost_app |
Get active application info |
launch_app |
Launch an application |
quit_app |
Quit an application |
hide_app |
Hide an application |
is_app_running |
Check if app is running |
Notifications (3 tools)
| Tool | Description |
|---|---|
send_notification |
Send macOS notification |
send_alert |
Show alert dialog |
choose_from_list |
Show selection dialog |
Clipboard (3 tools)
| Tool | Description |
|---|---|
get_clipboard |
Get clipboard text |
set_clipboard |
Copy text to clipboard |
clear_clipboard |
Clear clipboard |
Finder (6 tools)
| Tool | Description |
|---|---|
reveal_in_finder |
Reveal file in Finder |
create_folder |
Create new folder |
move_to_trash |
Move items to Trash |
empty_trash |
Empty Trash |
get_selected_files |
Get Finder selection |
search_files |
Search using Spotlight |
Calendar (3 tools)
| Tool | Description |
|---|---|
list_calendars |
List all calendars |
list_calendar_events |
List upcoming events |
create_calendar_event |
Create new event |
Reminders (5 tools)
| Tool | Description |
|---|---|
list_reminder_lists |
List reminder lists |
list_reminders |
List reminders |
create_reminder |
Create new reminder |
complete_reminder |
Mark reminder complete |
get_overdue_reminders |
Get overdue items |
Scripts (4 tools)
| Tool | Description |
|---|---|
run_applescript |
Execute AppleScript code |
run_shell_command |
Execute shell command (with safety checks) |
open_url |
Open URL in browser |
open_file |
Open file with app |
Task Scheduler (3 tools)
| Tool | Description |
|---|---|
create_scheduled_task |
Create launchd task |
list_scheduled_tasks |
List managed tasks |
remove_scheduled_task |
Remove a task |
Complete Tool List (Alphabetical)
choose_from_list, clear_clipboard, complete_reminder, create_calendar_event,
create_folder, create_reminder, create_scheduled_task, empty_trash,
get_clipboard, get_frontmost_app, get_overdue_reminders, get_selected_files,
get_system_info, get_volume, get_wifi_network, hide_app, is_app_running,
is_dark_mode, launch_app, list_calendar_events, list_calendars,
list_reminder_lists, list_reminders, list_running_apps, list_scheduled_tasks,
move_to_trash, open_file, open_url, quit_app, remove_scheduled_task,
reveal_in_finder, run_applescript, run_shell_command, say_text, search_files,
send_alert, send_notification, set_clipboard, set_dark_mode, set_volume,
sleep_display, take_screenshot, toggle_dark_mode, toggle_mute, toggle_wifi
Example Usage
Via Claude/Cursor
"Set my system volume to 50%"
"What applications are currently running?"
"Send me a notification with title 'Reminder' and message 'Check email'"
"Create a calendar event for tomorrow at 2pm called 'Team Meeting'"
"What's on my clipboard?"
Via MCP Tool Calls
# Set volume
result = await callMCPTool('macos-automation__set_volume', {'level': 50})
# Send notification
result = await callMCPTool('macos-automation__send_notification', {
'title': 'Hello',
'message': 'World'
})
# List running apps
result = await callMCPTool('macos-automation__list_running_apps', {})
Security
Shell Command Safety
The run_shell_command tool includes safety checks that block dangerous commands. Only whitelisted commands are allowed:
- System info:
sw_vers,uname,hostname,whoami, etc. - File info:
ls,pwd,stat,file, etc. - Network info:
networksetup,ifconfig, etc. - macOS utilities:
defaults,launchctl,open, etc.
Dangerous patterns like rm -rf /, sudo rm, fork bombs, etc. are blocked.
AppleScript Sandboxing
AppleScript execution respects macOS sandboxing and permission systems. Scripts cannot access protected resources without proper permissions.
Development
Running Tests
uv run pytest tests/ -v
Verification
# Verify server loads correctly
cd /Volumes/dev-4tb/AA-GitHub/MCP/mcp-applescript-automation
uv run python -c "from macos_automation_mcp import mcp; print(f'Server: {mcp.name}')"
# List all registered tools
uv run python -c "
from macos_automation_mcp.server import mcp
tools = list(mcp._tool_manager._tools.keys())
print(f'Registered {len(tools)} tools')
for t in sorted(tools): print(f' - {t}')
"
# Test basic functionality
uv run python -c "
from macos_automation_mcp.domains import system
result = system.get_system_info()
print(f'macOS: {result.get(\"macos_version\")}')
print(f'Host: {result.get(\"hostname\")}')
"
Project Structure
mcp-applescript-automation/
├── macos_automation_mcp/
│ ├── __init__.py
│ ├── server.py # FastMCP server with 45 tool definitions
│ ├── core/
│ │ ├── __init__.py
│ │ ├── applescript.py # AppleScript execution engine (osascript)
│ │ ├── shell.py # Safe shell command execution
│ │ └── scheduler.py # launchd task integration
│ ├── domains/
│ │ ├── __init__.py
│ │ ├── system.py # System controls (12 tools)
│ │ ├── applications.py # Application control (6 tools)
│ │ ├── finder.py # Finder operations (6 tools)
│ │ ├── calendar.py # Calendar.app integration (3 tools)
│ │ ├── reminders.py # Reminders.app integration (5 tools)
│ │ ├── notifications.py # Notification Center (3 tools)
│ │ ├── clipboard.py # Clipboard operations (3 tools)
│ │ └── scripts.py # Script execution (4 tools)
│ └── models/
│ ├── __init__.py
│ └── schemas.py # Pydantic models for type safety
├── tests/
│ ├── __init__.py
│ └── test_applescript.py # Unit tests
├── main.py # Entry point
├── pyproject.toml # Dependencies: mcp[cli], pydantic
├── uv.lock # Locked dependencies
└── README.md
Dependencies
- mcp[cli]>=1.0.0 - FastMCP framework with CLI support
- pydantic>=2.0.0 - Type-safe data validation
License
MIT License
Contributing
Contributions are welcome! Please feel free to submit issues and pull requests.
Acknowledgments
- Built using FastMCP
- Inspired by applescript-mcp
- Uses the Model Context Protocol
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.