AI Assistant

A personal AI assistant that intelligently manages your calendar events and emails using advanced language models and Google APIs integration.

Purpose

AI Assistant is designed to streamline your productivity by providing natural language interaction for:

• Calendar Management: Create one-time and recurring events with intelligent date parsing
• Email Management: Send emails directly from natural language commands
• Smart Scheduling: Understand relative dates like "next Tuesday" or "tomorrow" and convert them to proper calendar entries
• File system: Show content of specified directory, open files, and remove files/folders

The assistant uses an agentic architecture with LangChain and LangGraph to process user commands and execute tools intelligently, making your scheduling and communication effortless.

The Agent

AI Assistant is built using LangChain and LangGraph, which enables:

• Agentic workflow: The AI agent can decide which tools to use based on user input
• Tool calling: Seamlessly integrates with Google Calendar and Gmail APIs
• In-memory checkpointing: Maintains conversation state across interactions
• Error handling: Gracefully manages tool execution failures

The agent uses Ollama for running the language model locally, ensuring privacy and eliminating the need for API keys to external LLM providers.

Tools & Features Available

Calendar Tools

• add_event_to_calendar: Create single events with customizable reminders

  • Parameters: event name, location, description, start/end times, timezone, email/popup reminders
  • Example: "Add a meeting tomorrow at 3 PM"

• add_recurrent_event_to_calendar: Create recurring events (daily, weekly, monthly)

  • Supports complex recurrence rules (e.g., "every Tuesday and Thursday")
  • Example: "Create a weekly standup every Monday at 10 AM"

• get_calendar_events: Retrieve upcoming events

  • Useful for checking availability before scheduling

Email Tools

• send_email: Send emails with customizable subject and body
  • Parameters: recipient email, subject, body
  • Example: "Send an email to john@example.com about the project deadline"

File System Tools

• list_directory: Browse files and folders in a directory

  • Example: "List files in my documents folder"

• read_file: Read and display file contents

  • Example: "Read my notes from the latest meeting"

• delete_file: Remove files from the system

  • Example: "Delete the old backup file"

⚠️ Administrator Rights Required: To access protected system files and directories, AI Assistant must be run with administrator privileges. On Windows, right-click on your terminal/PowerShell and select "Run as administrator" before launching the application.

Utility Tools

• get_current_time: Retrieve the current date and time for accurate scheduling

Installation Guide

Prerequisites

• Python 3.8 or higher
• Git
• Administrator access (for Ollama installation on Windows)

1. Install Ollama

Ollama allows you to run large language models locally on your machine.

Windows:

1. Download the installer from ollama.ai
2. Run the installer and follow the setup wizard
3. Ollama will install and start automatically

macOS:

curl -fsSL https://ollama.ai/install.sh | sh

Linux:

curl -fsSL https://ollama.ai/install.sh | sh

2. Pull the Language Model

This project uses Qwen3:8b, a lightweight yet powerful language model that offers excellent performance with minimal resource requirements:

ollama pull qwen3:8b

Why Qwen3:8b

• Low weight: Only 5GB, suitable for most systems
• Fast inference: Quick response times for an interactive assistant
• Strong performance: Excellent understanding of instructions and context
• Open source: No external API dependency

Note: You're free to use any other Ollama-supported model. Simply replace qwen3:8b with your preferred model (e.g., neural-chat, llama2, dolphin-mixtral) in the ai_chatbot.py configuration.

3. Clone and Setup the Repository

# Clone the repository
git clone https://github.com/ngiani/AI-Assitant.git
cd AI-Assistant

# Create a virtual environment
python -m venv venv

# Activate the virtual environment
# On Windows:
venv\Scripts\activate
# On macOS/Linux:
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

4. Setup Google Authentication

AI-Assistant integrates with Google Calendar and Gmail. You need to authorize it:

1. Create a Google Cloud Project:

   • Go to Google Cloud Console
   • Create a new project (e.g., "AI-Assistant")
   • Enable the following APIs:
     • Google Calendar API
     • Gmail API

2. Create OAuth 2.0 Credentials:

   • In the Cloud Console, go to "Credentials"
   • Click "Create Credentials" → "OAuth client ID"
   • Choose "Desktop application"
   • Download the credentials as JSON

3. Add credentials.json:

   • Place the downloaded credentials.json file in the project root directory
   • The app will automatically use it for authentication

4. Setup Environment Variables:

   • Create a .env file in the project root
   • Add your email address:

     EMAIL_ADDRESS=your.email@gmail.com
     

   • This file is automatically ignored by git to protect your privacy

5. Run the Application

# Make sure Ollama is running
ollama serve

# In another terminal, activate venv and run:
python ai_chatbot.py

Configuration Files

• credentials.json: Your Google OAuth credentials (created in setup step 4)
• .env: Environment variables including your email address for sending emails (to be added by the user)
• system_prompt.txt: System instructions for AI-Assitant's behavior
• requirements.txt: Python package dependencies

Google API

The application automatically handles Google API tokens:

• calendar_token.json: Auto-generated after first calendar access
• gmail_token.json: Auto-generated after first email access

These files are in .gitignore to protect your privacy.

Usage Examples

"Add an event called 'Team Meeting' next Tuesday at 2 PM"
"Create a weekly standup every Monday and Wednesday at 10 AM with a 1-day reminder"
"Send an email to alice@example.com with subject 'Project Update' and the latest status"
"What events do I have next week?"
"Remind me about the conference in 3 days"

Project Structure

AI-Assistant/
├── agent.py              # Agent implementation using LangChain
├── tools.py              # Calendar and email tools
├── ai_chatbot.py         # Main chatbot interface
├── system_prompt.txt     # Instructions for the AI agent
├── requirements.txt      # Python dependencies
├── credentials.json      # Google OAuth credentials (user-added)
├── .env                  # Environment variables including EMAIL_ADDRESS (user-added)
└── README.md             # This file

Troubleshooting

Ollama not connecting:

• Ensure Ollama is running: ollama serve
• Check if port 11434 is accessible

Google API errors:

• Delete the token files (calendar_token.json, gmail_token.json) and re-authenticate
• Ensure credentials.json is in the project root

Email not sending:

• Verify that the .env file exists and contains EMAIL_ADDRESS=your.email@gmail.com
• Ensure the email address is authorized in Google account settings

Model not found:

• Run ollama pull <model> to download the model
• Check available models with ollama list

License

This project is open source and available under the MIT License.

Contributing

Feel free to fork, modify, and improve AI-Assistant for your needs!