keyboard-maestro-mcp
Enables managing Keyboard Maestro macros with natural language commands, including creation, editing, execution, and debugging.
README
Keyboard Maestro MCP
An MCP server for Keyboard Maestro. Manage your macros with natural language commands.
What Can You Do?
Instead of digging through menus and clicking buttons, just ask:
| Just say... |
|---|
| "Run my morning routine macro" |
| "Create a macro that types my email signature when I press Cmd+Shift+S" |
| "What macros have been failing lately? Fix them." |
| "Duplicate my 'Open Project' macro and modify it for the new client" |
| "Disable all my work macros, I'm on vacation" |
Your AI assistant becomes a power user of your Mac. It can read, create, modify, run, and debug your Keyboard Maestro macros through conversation.
Real-World Examples
| Use Case | Example |
|---|---|
| Debugging | "My clipboard manager macro stopped working yesterday. Check the logs and tell me what went wrong." |
| Bulk Editing | "I have 20 macros that open Chrome. Change them all to open Arc instead." |
| Quick Automation | "Make a macro that mounts my NAS drives when I connect to my home WiFi." |
| Organization | "Create a new macro group called 'Client Work' and move all macros with 'ACME' in the name into it." |
[!IMPORTANT] This MCP gives your AI full access to Keyboard Maestro. It can run, modify, and delete macros. Always review suggested changes before approving them.
Requirements
- macOS (Keyboard Maestro is macOS-only)
- Keyboard Maestro installed and running
- Node.js v18+
Quick Start
git clone https://github.com/saihgupr/keyboard-maestro-mcp.git
cd keyboard-maestro-mcp
npm install
npm run setup
The setup wizard will configure your AI client automatically. It supports:
- Claude Desktop
- Cursor
- VS Code (Copilot)
- Windsurf
- Antigravity
- Any MCP-compatible client
Available Tools
Macro Management
| Tool | What it does |
|---|---|
km_search_macros |
Find macros by name |
km_list_macros |
List all macros |
km_get_macro |
Get macro details |
km_get_macro_xml |
Export a macro's full definition |
km_create_macro |
Build a new macro |
km_clone_macro |
Duplicate an existing macro |
km_delete_macro |
Remove a macro |
km_enable_macro |
Turn macros on/off |
km_run_macro |
Execute a macro right now |
km_manage_group |
Organize macro groups |
Action Editing
| Tool | What it does |
|---|---|
km_list_actions |
See all steps in a macro |
km_add_action |
Add a new step |
km_move_action |
Reorder steps |
km_delete_action |
Remove a step |
km_get_action_xml |
Export a step's definition |
km_set_action_xml |
Replace a step |
km_search_replace_action |
Find and replace text in a step |
Trigger Editing
| Tool | What it does |
|---|---|
km_add_trigger |
Add a new trigger |
km_delete_trigger |
Remove a trigger |
km_get_trigger_xml |
Export a trigger |
km_set_trigger_xml |
Replace a trigger |
Variables and Logs
| Tool | What it does |
|---|---|
km_manage_variable |
Read, write, or delete KM variables |
km_get_errors |
Find recent macro failures |
km_get_log |
Search the engine log |
Testing Your Setup
Quick check:
npm run verify
Interactive testing:
npx @modelcontextprotocol/inspector node build/index.js
Manual Configuration
If you prefer to configure manually, add this to your AI client's MCP config:
{
"mcpServers": {
"keyboard-maestro": {
"command": "node",
"args": ["/path/to/keyboard-maestro-mcp/build/index.js"]
}
}
}
Support
Having issues? Open an issue on GitHub
If you like this project, please consider giving the repo a ⭐ star!
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。