AISecretary
A modular collection of MCP servers for automating virtual secretary tasks within Microsoft Outlook, including email management, calendar operations, contacts, tasks, and mailbox settings through Microsoft Graph API.
README
AISecretary
AISecretary is a collection of MCP servers designed to automate the tasks of a virtual secretary insede the Outlook suite. Each server handles a specific tool, such as:
- Outlook Mail
- Outlook Calendar
- Outlook Contacts
- Outlook To Do
- Outlook Mailbox Settings
- Categories
Features
- Modular: Each tool runs as an independent server, so you can start only the ones you need.
- Easy integration: Compatible with OpenWebUI and other platforms supporting MCP.
- Extensible: Easily add new servers for additional tools or services.
- Automation: Handles repetitive tasks like checking emails, managing calendars, and more.
Prerequisites
- uv installed (Python environment and dependency manager).
- Python 3.11 or higher.
Getting Started
Follow the instructions below to configure and run the server:
Service Setup
To configure the necessary Microsoft Graph API settings, follow this setup guide:
Running the Server
Run the following command to start the server (replace server.py with your actual server file):
Recommended: Using Claude Desktop
If you have Claude Desktop installed:
uv run mcp install server.py
Other option to install your server in Claude Desktop is to edit the configuration file at:
C:\Users\YOUR_USER\AppData\Roaming\Claude\claude_desktop_config.json
{
"mcpServers": {
"Server_name": {
"command": "/Your/path/to/uv",
"args": [
"run",
"--with",
"mcp[cli]",
"--with",
"msal",
"mcp",
"run",
"/your/path/to/server.py"
]
},
}
}
If you want all the MCP servers you can just also run the script claude_setup.py wity uv.
uv run python claude_setup.py
Alternative: Running with OpenWebUI
If you're using platforms like OpenWebUI:
uvx mcpo --port 9000 -- uv run mcp run server.py
Then open OpenWebUI, go to:
Settings > Tools > [+]
Add the following URL:
http://127.0.0.1:9000
You can now interact with AISecretary through the OpenWebUI interface.
Functionalities
These are the available functionalities for each of the MCP servers:
Email Management
- Advanced email search
- Retrieve conversations
- Mark as read/unread
- Retrieve full emails with attachments
- Delete emails
- Move or copy emails
- Manage flags
Email Creation and Sending
- Create or edit drafts
- Handle attachments
- Send drafts
- Reply to emails
- Forward emails
Folder Organization
- View folder structure
- Create or edit folders
- Delete folders
- Navigate folder hierarchy
Mail Rules
- View existing rules
- Create or edit rules
- Delete rules
🏷️ Categories
- View existing categories
- Create or edit categories
- Delete categories
- Assign categories to emails
- Assign categories to events
- Use predefined colors
📅 Calendar
Event Management
- Retrieve events
- Create events
- Update events
- Delete events
- Retrieve detailed event information
Invitation Handling
- Accept invitations
- Decline invitations
- Respond tentatively
- Cancel events
Attachment Management
- Add attachments
- Remove attachments
Calendar Management
- List calendars
- Retrieve specific calendars
- Create calendars
- Update calendars
- Delete calendars
Calendar Group Management
- List groups
- Create groups
- Update groups
- Delete groups
Availability Lookup
- Retrieve free/busy schedule
✅ Task Management
Task List Management
- Retrieve lists
- Create lists
- Delete lists
Task Management
- Retrieve tasks
- Retrieve specific task
- Create tasks
- Update tasks
- Delete tasks
👥 Contacts
Contact Folder Management
- Create folders
- Delete folders
- Search folders
Contact Management
- Search contacts
- Retrieve detailed contact information
- Create contacts
- Update contacts
- Delete contacts
⚙️ Mailbox Settings
General Settings Management
- Retrieve settings
- Update settings
Supported Settings
- Automatic replies
- Time zone
- Working hours
Adding New Servers
To add support for new tools, create a new Python file following the structure of the existing servers and register the functions you want to expose as MCP tools.
More Information
- Python original sdk: Explains how to use MCP with different models and tools.
- OpenWebUI: Detailed guide on integrating MCP servers with OpenWebUI.
- Anthropic: Detailed guide on integrating MCP servers with Claude Desktop.
License
This project is licensed under the MIT License. See the LICENSE file for details.
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。