Project: Personal Day to Day AI Assistant

[DevOpsAIguru123]

Project Name

Personal Day to Day AI Assistant

Description

Why This Project?

The modern individual juggles a growing number of digital tools for managing everyday tasks—from banking to scheduling to information retrieval. However, these systems often operate in isolation, creating inefficiencies, context-switching, and fragmented user experiences.

AI Personal Assistant aims to solve this by unifying critical daily workflows into a single, intelligent interface powered by Azure OpenAI. By integrating with APIs like Google Calendar, Google Tasks, AccuWeather, and Google Custom Search, and combining them with a vectorized bank statement search feature, this assistant can:

• Save time by streamlining task management and scheduling
• Improve productivity by offering natural language-based interaction
• Provide quick, actionable insights from personal financial data
• Keep users informed with real-time weather and web information

Ultimately, this assistant empowers users to focus more on meaningful activities and less on navigating disjointed tools.

AI Personal Assistant

A web-based AI assistant that integrates with Azure OpenAI and various APIs to help with tasks like:

• Bank statement search and vector database querying
• Google Calendar management
• Google Tasks management
• Weather information via AccuWeather API
• Web searches via Google Custom Search API

Application Screenshots

Note: The placeholder images in the screenshots directory should be replaced with actual screenshots of your running application. Take screenshots of each functionality as you use the app and save them with the corresponding filenames.

Welcome Screen

The main welcome screen showing available capabilities of the assistant.

Bank Statement Information

Example of bank statement information retrieval showing balance and transaction details.

To-Do Task Management

View and manage your to-do tasks using natural language commands.

Weather Information

Current weather information and forecast with useful recommendations.

Web Search and News

Latest news and information retrieved from the web based on your queries.

Calendar Events

View and manage your calendar events, meetings, and appointments with natural language commands.

Features

Bank Statement Search

Search through your bank statements for transactions and financial information using Azure Cognitive Search with vector embeddings.

Google Calendar

View and manage your calendar events and appointments using Google Calendar API.

Google Tasks

Create, manage, and delete tasks and to-do lists using Google Tasks API.

Commands:

• Create new tasks: "Create a task to buy groceries"
• List all tasks: "Show me my tasks"
• Delete tasks: "Delete my task about buying groceries"

Weather Information

Get current weather conditions and forecasts for any location using the AccuWeather API.

Web Search

Search the web for general knowledge and current events using the Google Custom Search API.

Usage Examples

• "What's the weather like in New York?"
• "Show me my calendar for tomorrow"
• "Create a new task to buy groceries"
• "Find transactions from Starbucks in my bank statements"
• "Search for information about Mars"
• "What do my bank statements show about spending on restaurants?"

File Structure

• chainlit_app.py - Main application with Chainlit web interface
• open-ai.py - Azure OpenAI integration
• vector.py - Azure Cognitive Search vector database functions
• weather.py - AccuWeather API integration
• search_api.py - Google Search API integration
• tasks.py - Google Tasks API integration
• calander.py - Google Calendar API integration

Authentication

For Google services (Calendar and Tasks), the application uses OAuth 2.0 authentication. The first time you use these services, you'll need to authorize the application by following the OAuth flow.

Security Notes

All API keys and secrets are stored in environment variables for security. Never commit your .env file to version control.

Language & Framework

• Python
• C#
• Java
• JavaScript/TypeScript
• Microsoft Copilot Studio
• Microsoft 365 Agents SDK
• Azure AI Agent Service

Project Repository URL

https://github.com/DevOpsAIguru123/ms-agent-hack-2025

Deployed Endpoint URL

No response

Project Video

https://www.loom.com/share/020cbb29bc3e4bbfa144f48db6e152e2?sid=0b313527-ee87-419c-a90a-7877b4f58fb2

Team Members

only myself

Registration Check

• Each of my team members has filled out the registration form

[DevOpsAIguru123]

@multispark please refer here #530

[cole-g-johnson]

Thank you for participating in Microsoft's AI Agents Hackathon 2025! Please have each team member fill out this form to collect your digital badges: https://aka.ms/agentshack/badges