remindctl-mcp
MCP server for Apple Reminders via remindctl. Enables MCP-compatible AI assistants to read and write Apple Reminders through the standard Model Context Protocol.
README
remindctl-mcp
MCP server for Apple Reminders via remindctl.
Lets any MCP-compatible AI assistant (Amazon Quick, Claude Desktop, etc.) read and write your Apple Reminders through the standard Model Context Protocol.
Requirements
- macOS 14+ (Sonoma or later)
- remindctl installed (
brew install steipete/tap/remindctl) - Python 3.11+
- uv (recommended) or pip
Setup (Amazon Quick)
1. Clone & install dependencies
git clone https://github.com/ebi-yade/remindctl-mcp.git
cd ./remindctl-mcp
uv venv
uv pip install mcp
2. Register as MCP server
Open Settings → Capabilities → Connectors → + Add MCP server and configure:
| Field | Value |
|---|---|
| Name | remindctl |
| Command | /path/to/remindctl-mcp/.venv/bin/python |
| Args | /path/to/remindctl-mcp/server.py |
Replace /path/to/ with the actual location (e.g. /Users/you/remindctl-mcp/).
3. Grant Reminders permission
macOS controls Reminders access via TCC. Permission is granted to the host app that spawns the MCP server process (e.g. Amazon Quick), not to python or remindctl directly.
- Ask the assistant to call
reminders_authorize - A macOS dialog will appear — click OK
- Verify with
reminders_status→ should show"authorized": true
TCC Troubleshooting
If any tool returns "Reminders access denied", the fix is always the same:
# 1. Reset permission to trigger a fresh prompt
tccutil reset Reminders <bundle_id>
# For Amazon Quick:
tccutil reset Reminders com.amazon.QuickWork.mac
Then ask the assistant to call reminders_authorize again and click OK on the dialog.
Why this happens: macOS may revoke TCC grants after OS updates, or the user may have accidentally denied the prompt.
tccutil resetclears the decision so the dialog can appear again.
Tools
| Tool | Description |
|---|---|
reminders_authorize |
Trigger TCC permission dialog (for initial setup / recovery) |
reminders_status |
Check authorization status |
reminders_show |
Show reminders by filter (today, week, overdue, etc.) |
reminders_search |
Search reminders by keyword |
reminders_lists |
List all reminder lists or show reminders in specific list(s) |
reminders_add |
Create a new reminder |
reminders_edit |
Edit an existing reminder |
reminders_complete |
Mark reminders as complete |
reminders_delete |
Delete reminders |
How it works
MCP Host (Amazon Quick, Claude, etc.)
↕ stdio (JSON-RPC)
remindctl-mcp (this server)
↕ subprocess
remindctl CLI
↕ EventKit API
Apple Reminders (syncs via iCloud)
The server is a thin wrapper: it translates MCP tool calls into remindctl CLI invocations and returns JSON output. No data is stored or cached — all state lives in Apple Reminders.
Other MCP hosts
This server uses stdio transport and works with any MCP client. For Claude Desktop, add to claude_desktop_config.json:
{
"mcpServers": {
"remindctl": {
"command": "/path/to/remindctl-mcp/.venv/bin/python",
"args": ["/path/to/remindctl-mcp/server.py"]
}
}
}
License
MIT
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。
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。
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。
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。
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。
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。
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。
mcp-server-qdrant
这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。