mcp-mac
Enables AI agents to interact with macOS applications (Finder, Mail, Contacts, Reminders, Notes, Calendar, TextEdit) using AppleScript. Allows AI assistants to perform tasks like searching contacts, managing files, checking email, and creating reminders through natural language.
README
mcp-mac
AI-powered automation of macOS Apple applications
This is a proof of concept to learn about AI and MacOS, and developed using with AI. I am unlikely to progress this any further.
mcp-mac enables AI agents to interact with macOS applications (Finder, Mail, Contacts, etc) using AppleScript through the Model Context Protocol (MCP). This allows AI assistants to perform tasks like searching contacts, managing files, and checking email on your behalf.
What is MCP?
The Model Context Protocol (MCP) is an open standard that allows AI applications to securely connect to external data sources and tools. mcp-mac implements an MCP server that provides AI agents with controlled access to your macOS applications.
Current Status
Available Tools
- Finder: Get selected files, get current folder location
- Mail: List accounts, get inbox unread/total count, create and send messages
- Contacts: Search people, get contact information
- Reminders: List lists, get incomplete reminders, create reminders
- Notes: List folders, search/get recent notes, create notes
- TextEdit: List documents, create documents
- Calendar: List calendars, get today's/upcoming events, create events
- Workflow: Composite tools for multi-app actions (e.g., Contacts to TextEdit)
- Discovery: Introspect application capabilities
Quick Start
Prerequisites
- macOS (tested on MacOS 26.2)
- Node.js 18
Installation
# Clone the repository
git clone <repository-url>
cd mcp-mac
# Install dependencies
npm install
# Build the project
npm run build
E.g usage with Q CLI
# Add to your Q CLI MCP configuration
q config mcp add mcp-mac ./dist/index.js
# Use in Q chat sessions
# The tools will be available as mcp-mac___tool-name
Example Workflow
Once set up, you can ask an AI assistant to:
"Get John Smith's contact information and also check how many unread emails I have"
The AI will use the mcp-mac tools to:
- Search for "John Smith" in Contacts
- Retrieve his contact details
- Check your Mail inbox count
- Provide a combined response
Project Structure
mcp-mac/
├── src/index.ts # MCP server implementation
├── dist/ # Compiled JavaScript
├── docs/ # Project documentation
│ ├── PROJECT-CHARTER.md # Project governance
│ ├── DEVELOPMENT-GUIDE.md # Development processes
│ └── iterations/ # Implementation history
└── package.json # Dependencies and scripts
Documentation
- Architecture - System design and technical details
- Technical Lessons - AppleScript quirks and implementation details
Privacy & Security
mcp-mac accesses your personal data (contacts, email, files) through standard macOS AppleScript interfaces. All data processing happens locally on your machine. No data is transmitted to external servers.
Important: Only use with trusted AI applications and review any automation requests before execution.
Disclaimer
This software is provided "as is," without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose and non-infringement. The entire risk as to the quality and performance of the program is with you. In no event shall the author or copyright holders be liable for any claim, damages, or other liability, whether in an action of contract, tort or otherwise, arising from, out of or in connection with the software or the use or other dealings in the software. You assume all responsibility for the use of this software on your system, and you are accountable for any actions or consequences that arise from its use.
License
Licensed under the MIT license. MIT
Copyright
Copyright (c) Sami Cokar, 2025
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。