Email Analyzer

Email Analyzer is a sophisticated email analysis application that leverages AI to provide deep insights into your email communications. Built with .NET 8 and utilizing OpenAI's advanced language models, it offers semantic search, content summarization, and intelligent analysis of email patterns.

Project Structure

The solution is divided into three main projects:

• EmailAnalyzer.Server: Backend ASP.NET Core Web API service, deployed on Railway
• EmailAnalyzer.Client: Frontend .NET MAUI application
• EmailAnalyzer.Shared: Common models, interfaces and services shared between Client and Server

Deployment

• Backend service is hosted on Railway platform
• MongoDB database is also hosted on Railway platform
• Frontend is distributed as a standalone MAUI application

Features

• Multi-Provider Support: Seamless integration with Gmail and Outlook
• OAuth2 Authentication: Secure authentication flow for email providers
• Semantic Search: AI-powered search using embeddings
• Smart Summarization: Intelligent email content analysis and summarization
• Vector Search: Similarity-based email search using embeddings
• Real-time Analysis: Dynamic processing of email content
• Cross-Platform UI: Built with .NET MAUI for wide platform compatibility

Architecture

The application follows a clean architecture pattern and consists of three main components:

Backend (.NET 8 Web API)

• REST API endpoints for email operations
• Integration with OpenAI API
• MongoDB for storing emails and embeddings
• Token management and OAuth2 flow

Backend Flow for Authentication:

sequenceDiagram
    participant User
    participant Client
    participant Server
    participant Gmail
    participant Outlook
    User ->>+ Client: Request Authorization URL
    Client ->>+ Server: Get Authorization URL (e.g., Gmail/Outlook)
    alt Gmail Provider
        Server ->>+ Gmail: Generate Auth URL
        Gmail -->>- Server: Auth URL with state=gmail
    else Outlook Provider
        Server ->>+ Outlook: Generate Auth URL
        Outlook -->>- Server: Auth URL with state=outlook
    end
    Server -->>- Client: Auth URL
    Client -->> User: Redirect to Auth URL

    alt User Authenticates with Gmail
        User ->>+ Gmail: Authenticate (e.g., login & consent)
        Gmail -->>- Client: Redirect with code and state=gmail
    else User Authenticates with Outlook
        User ->>+ Outlook: Authenticate (e.g., login & consent)
        Outlook -->>- Client: Redirect with code and state=outlook
    end

    Client ->>+ Server: Callback with Authorization Code and State
    alt Gmail Provider
        Server ->>+ Gmail: Exchange Code for Tokens
        Gmail -->>- Server: Access & Refresh Tokens
    else Outlook Provider
        Server ->>+ Outlook: Exchange Code for Tokens
        Outlook -->>- Server: Access & Refresh Tokens
    end

    Server ->>+ TokenStorage: Store Tokens
    TokenStorage -->>- Server: Tokens Stored
    Server -->>- Client: Authentication Success

Loading

Backend Flow for Email Analysis:

sequenceDiagram
    title Email Analyzer Backend Flow
    participant Client
    participant EmailController
    participant TokenStorage
    participant EmailService
    participant OpenAIService
    participant MongoDB
%% Authentication Flow
    Client ->> EmailController: Authentication Request
    EmailController ->> TokenStorage: Store Provider Tokens
    TokenStorage -->> EmailController: Token Stored
    EmailController -->> Client: Auth Success
%% Email Fetching & Analysis
    Client ->> EmailController: Request Email Analysis
    EmailController ->> TokenStorage: Get Provider Token
    TokenStorage -->> EmailController: Return Token
    EmailController ->> EmailService: Fetch Emails
    EmailService -->> EmailController: Return Email Data

    par Process Emails
        EmailController ->> OpenAIService: Generate Embeddings
        OpenAIService -->> EmailController: Return Embeddings
        EmailController ->> MongoDB: Store Emails with Embeddings
        MongoDB -->> EmailController: Confirm Storage
    and Analyze Content
        EmailController ->> OpenAIService: Generate Summary
        OpenAIService -->> EmailController: Return Analysis
    end

    EmailController -->> Client: Return Analysis Results
%% Semantic Search
    Client ->> EmailController: Semantic Search Query
    EmailController ->> OpenAIService: Generate Query Embedding
    OpenAIService -->> EmailController: Return Query Embedding
    EmailController ->> MongoDB: Search Similar Documents
    MongoDB -->> EmailController: Return Similar Emails
    EmailController ->> OpenAIService: Analyze Search Results
    OpenAIService -->> EmailController: Return Search Analysis
    EmailController -->> Client: Return Search Results & Analysis

Loading

Frontend (.NET MAUI)

• Cross-platform UI implementation
• Responsive design
• Real-time data visualization
• Search interface

Shared Library

• Common models and interfaces
• DTOs for data transfer
• Shared utilities and helpers

Prerequisites

• .NET 8 SDK
• MongoDB
• OpenAI API Key
• Gmail/Outlook API credentials

Configuration

1. Backend Configuration (appsettings.json):

{
  "MongoDB": {
    "ConnectionString": "your_connection_string",
    "DatabaseName": "emailanalyzer"
  },
  "OpenAI": {
    "ApiKey": "your_api_key",
    "EmbeddingModel": "text-embedding-ada-002",
    "CompletionModel": "gpt-4"
  }
}

2. Email Provider Configuration:

{
  "Gmail": {
    "ClientId": "your_client_id",
    "ClientSecret": "your_client_secret",
    "RedirectUri": "your_redirect_uri"
  },
  "Outlook": {
    "ClientId": "your_client_id",
    "ClientSecret": "your_client_secret",
    "RedirectUri": "your_redirect_uri"
  }
}

Installation

1. Clone the repository:

git clone https://github.com/yourusername/email-analyzer.git

2. Install dependencies:

dotnet clean
dotnet restore
dotnet build

3. Configuration:

• Set up provider credentials in server configuration
• Configure OpenAI API key
• Set client API base URL

4. Run the application:

# Run MAUI client
cd EmailAnalyzer.Client
dotnet run

Note: Since both backend and database are hosted on Railway, no local database setup is required.

API Endpoints

Authentication

• GET /api/auth/url/{provider} - Get OAuth2 authorization URL
• POST /api/auth/authenticate - Handle OAuth2 authentication
• GET /api/auth/callback - OAuth2 callback handling
• GET /api/auth/all-tokens - Get all stored tokens
• DELETE /api/auth/remove-token/{provider} - Remove stored token

Email Operations

• GET /api/email/test/{provider} - Test provider connection
• POST /api/email/{provider} - Fetch emails for date range
• GET /api/email/available-range/{provider} - Get available date range
• POST /api/email/{provider}/analyze - Analyze emails for date range
• GET /api/email/{provider}/summaries - Get email summaries
• GET /api/email/{provider}/search - Semantic search with AI analysis
• GET /api/email/mongo - Check MongoDB connection

Health Checks

• GET /api/health - Basic service health check
• GET /api/health/mongo - MongoDB connection health check

Development

Building for Production

# Build backend
dotnet publish -c Release

# Build MAUI client
dotnet publish -f net8.0-windows10.0.19041.0

Security

• OAuth2 flow for secure authentication
• Secure token storage
• API key protection
• Rate limiting
• Input validation

Contributing

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Create a Pull Request

License

This project is licensed under the MIT License - see the LICENSE.md file for details.

Acknowledgments

• OpenAI for providing the AI models
• MongoDB for database capabilities
• Microsoft for .NET MAUI framework