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.

Functionalities

These are the available functionalities for each of the MCP servers:

📧 Email

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.

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:

• Outlook

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.