Email Tracking System

A smart email processing system powered by LangGraph that automatically categorizes incoming emails, extracts job application details, and provides notifications for urgent messages. Features a Streamlit dashboard for interactive email management.

Features

• Email Download: Automatically downloads emails from your connected account
• Email Classification: Categorizes emails into spam, job, urgent, and general
• Job Application Tracking: Extracts and organizes job application details
• Desktop Notifications: Alerts you about urgent emails
• Concurrent Processing: Processes different email categories in parallel
• Streamlit Dashboard: Interactive web interface for monitoring and managing emails
• LangGraph Workflow: Structured processing pipeline with visualization
• Agent-Based Architecture: Modular design with specialized agents for different tasks

Architecture Diagram

Project Structure

email-tracking-system/
├── agents/              # Specialized processing agents
│   ├── classification_agent.py  # Email classification agent
│   ├── summarization_agent.py   # Email summarization agent
│   └── __init__.py
├── core/                # Core system functionality
│   ├── email_downloader.py  # Downloads emails from server
│   ├── email_fetcher.py     # Fetches emails via IMAP
│   ├── initialize_db.py     # Database initialization
│   ├── utils.py            # Utility functions
│   └── __init__.py
├── models/              # Data models
│   ├── email.py         # Email and State type definitions
│   └── __init__.py
├── processing/          # Email processing modules
│   ├── fetch_emails.py    # Fetches unprocessed emails
│   ├── process_all.py     # Concurrent processing of all categories
│   ├── process_general.py # General email processing
│   ├── process_jobs.py    # Job email processing
│   ├── process_spam.py    # Spam email processing
│   ├── process_urgent.py  # Urgent email processing
│   └── __init__.py
├── prompts/             # LLM prompt templates
│   ├── classification_prompt.py # Email classification prompts
│   ├── job_extraction.py       # Job detail extraction prompts
│   ├── summarization_prompt.py # Email summarization prompts
│   └── __init__.py
├── workflows/          # Processing workflows
│   ├── graph_builder.py  # LangGraph workflow definition
│   └── __init__.py
├── jobs/               # Job application management
│   ├── applications.py   # Handle job applications
│   ├── tracker.py        # Track job applications
│   └── viewer.py         # View job applications
├── .streamlit/         # Streamlit configuration
├── app.py              # Streamlit web application
├── main.py             # CLI application entry point
├── config.py           # System configuration
├── apply_for_job.py    # Job application script
├── view_applications.py # View job applications script
├── emails.db           # SQLite database for email storage
├── jobs.xlsx           # Excel file for job application tracking
├── requirements.txt    # Project dependencies
└── README.md           # This file

Setup

1. Clone the repository:

git clone https://github.com/250519/Email-Tracking-Agent----Langgraph.git
cd email-tracking-system-langgraph

2. Create a virtual environment and install dependencies:

python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
pip install -r requirements.txt

3. Create a .env file with your email credentials:

IMAP_SERVER=imap.example.com
[email protected]
PASSWORD=your-password
OLLAMA_BASE_URL=http://localhost:11434  # If using Ollama

4. Run the application:

• For CLI version:

python main.py

• For web dashboard:

streamlit run app.py

Requirements

• Python 3.10+
• Required packages are listed in requirements.txt
• Ollama for running LLMs locally
• IMAP-enabled email account

How It Works

The system uses a LangGraph-powered workflow with specialized agents:

1. Email Download: Connects to your email server and downloads unread emails
2. Summarization Agent: Processes and summarizes each email to extract key information
3. Classification Agent: Analyzes emails and assigns them to one of four categories
4. Parallel Processing:
   • Spam: Marked as processed and categorized
   • Job: Details extracted and stored in jobs.xlsx
   • Urgent: Notification sent to desktop and marked as processed
   • General: Marked as processed

License

MIT License