AutoEmail AI

An Autonomous Email Intelligence Agent platform built to automate inbox management. AutoEmail AI uses a powerful pipeline of intelligent agents to read, classify, prioritize, and reply to your emails—saving you hours of manual work while keeping you in full control.

Features

• Email Reader Agent: Seamlessly connects to your inbox and fetches new emails in real-time.
• Intent Classification: Uses Natural Language Processing (NLP) to detect the exact intent of the sender (e.g., Support, Meeting, Sales).
• Priority Detection: Evaluates the urgency based on context, sentiment, and intent to ensure critical messages are handled first.
• Smart Reply Agent: Automatically drafts highly contextual, human-like responses perfectly tailored to the sender's tone.
• Approval Workflow: A safe "Human-in-the-loop" queue where you can review, approve, or reject AI-generated replies with one click.
• Modern SaaS Dashboard: A sleek, glassmorphism-styled dashboard (inspired by Linear and Stripe) to monitor your stats, view processing pipelines, and manage the approval queue.

The Agent Workflow Pipeline

AutoEmail AI operates on a fully transparent, 6-step pipeline:

1. Read: Connects to Gmail and fetches unread emails.
2. Detect Intent: Analyzes the email content to categorize it.
3. Prioritize: Assigns a priority level (High, Medium, Low).
4. Generate Reply: Drafts the perfect response.
5. Human Approval: Queues the email for your review on the dashboard.
6. Send: Dispatches the approved reply and verifies delivery in your Sent folder.

Project Structure

Autonomous Email Intelligence Agent/
├── backend/
│   ├── agents/               # Contains the AI logic (intent, priority, reply, etc.)
│   ├── main.py               # FastAPI backend server
│   ├── database.py           # SQLAlchemy database setup
│   ├── requirements.txt      # Python dependencies
│   ├── .env                  # Environment variables (Credentials)
│   └── venv/                 # Python Virtual Environment
├── frontend/
│   ├── index.html            # The main dashboard
│   ├── landing.html          # Modern SaaS landing page
│   ├── workflow.html         # Live agent workflow pipeline
│   ├── inbox.html            # Approval queue and email list
│   ├── app.js                # Frontend API logic and UI state
│   ├── style.css             # Main stylesheet
│   └── landing.css           # Glassmorphism aesthetic stylesheet
├── run_all.bat               # Easy 1-click startup script for Windows
├── .gitignore                # Git ignore rules
└── README.md                 # Project documentation

Getting Started

Prerequisites

• Python 3.9+
• A Gmail account with an App Password generated (for IMAP/SMTP access).

1. Configure Credentials

Navigate to the backend/ directory and create or update the .env file with your credentials:

EMAIL_ADDRESS=[email protected]
EMAIL_APP_PASSWORD=your_app_password

2. Run the Application (Windows)

We've included a convenient batch script to start both the backend API and the frontend server simultaneously. Simply double-click the run_all.bat file in the root directory.

Alternatively, to run manually: Backend:

cd backend
python -m venv venv
venv\Scripts\activate
pip install -r requirements.txt
uvicorn main:app --port 8002

Frontend:

cd frontend
python -m http.server 8080

3. Access the Platform

• Landing Page: http://localhost:8080/landing.html
• Dashboard: http://localhost:8080/
• Backend API Docs: http://localhost:8002/docs

🛠 Tech Stack

• Backend: Python, FastAPI, SQLAlchemy, LangGraph (optional streaming), SQLite.
• Frontend: Vanilla HTML5, CSS3, JavaScript (No heavy frameworks required).
• Design: Modern Glassmorphism, CSS Grid/Flexbox, custom keyframe animations.