MailPilot AI – AI-Powered Gmail Copilot

MailPilot AI is a full-stack Gmail automation dashboard that connects with a user's Gmail account, analyzes emails, generates smart replies, creates Gmail drafts, and stores AI analysis history using MongoDB.

It is designed as an AI-powered productivity tool for managing Gmail workflows through a clean dashboard, smart inbox, reply assistant, analysis history, and system health monitoring.

🚀 Project Overview

MailPilot AI helps users manage their Gmail inbox intelligently by combining Gmail API integration, AI-based email analysis, smart reply generation, and persistent history tracking.

The application allows users to:

Connect Gmail securely using Google OAuth
Fetch real Gmail emails
Analyze emails using an AI fallback engine
Classify emails by priority and category
Generate smart reply drafts
Create Gmail drafts directly from the dashboard
Save and view AI analysis history
Search, filter, and manage analyzed emails
Monitor backend, Gmail, AI mode, and MongoDB status

✨ Key Features

🔐 Gmail OAuth Integration

Secure Gmail authentication using Google OAuth 2.0 with token persistence.

📩 Smart Inbox

Fetches real Gmail emails and displays them in a searchable, filterable dashboard.

🤖 AI Email Analysis

Analyzes emails and generates:

Email summary
Priority level
Email category
Suggested response

✍️ AI Reply Draft Assistant

Generates editable reply drafts with tone control and allows users to create Gmail drafts directly.

🗂️ MongoDB Analysis History

Stores analyzed emails in MongoDB Atlas so users can revisit, search, filter, and delete past analysis records.

📊 Dashboard Analytics

Provides overview statistics such as loaded emails, analyzed emails, saved history, high-priority emails, and system status.

⚙️ System Health Page

Displays backend health, Gmail connection, AI mode, MongoDB history count, tech stack, and project module status.

🎨 Premium UI/UX

Includes a modern landing page, dark glass dashboard, animated visual elements, responsive layout, and clean product-style interface.

🛠️ Tech Stack

Frontend

React.js
Vite
CSS
Framer Motion
Lucide React
Axios

Backend

Node.js
Express.js
Gmail API
Google OAuth 2.0
MongoDB Atlas
Mongoose

AI Layer

Rule-based fallback AI engine
Email summarization
Priority detection
Reply generation

📁 Project Structure

mailpilot-ai/
│
├── backend/
│   ├── config/
│   │   └── db.js
│   ├── models/
│   │   ├── User.js
│   │   └── AnalyzedEmail.js
│   ├── routes/
│   │   ├── authRoutes.js
│   │   ├── gmailRoutes.js
│   │   └── aiRoutes.js
│   ├── services/
│   │   ├── googleAuthService.js
│   │   ├── gmailService.js
│   │   └── aiService.js
│   ├── utils/
│   │   └── emailUtils.js
│   ├── server.js
│   ├── package.json
│   └── .env.example
│
├── frontend/
│   ├── src/
│   │   ├── components/
│   │   │   └── Sidebar.jsx
│   │   ├── pages/
│   │   │   ├── LandingPage.jsx
│   │   │   └── Dashboard.jsx
│   │   ├── services/
│   │   │   └── api.js
│   │   ├── App.jsx
│   │   ├── App.css
│   │   └── main.jsx
│   ├── package.json
│   └── vite.config.js
│
├── .gitignore
└── README.md

⚙️ Environment Variables

Create a .env file inside the backend folder.

Use .env.example as reference:

PORT=5000
FRONTEND_URL=http://localhost:5173

GOOGLE_CLIENT_ID=your_google_client_id_here
GOOGLE_CLIENT_SECRET=your_google_client_secret_here
GOOGLE_REDIRECT_URI=http://localhost:5000/auth/google/callback

MONGO_URI=your_mongodb_atlas_connection_string_here

AI_MODE=fallback

Never upload your real .env file to GitHub.

▶️ How to Run Locally

1. Clone the repository

git clone https://github.com/javcod/mailpilot-ai.git
cd mailpilot-ai

2. Install backend dependencies

cd backend
npm install

3. Start backend server

npm run dev

Backend runs on:

http://localhost:5000

4. Install frontend dependencies

Open a new terminal:

cd frontend
npm install

5. Start frontend

npm run dev

Frontend runs on:

http://localhost:5173

🔗 Main API Endpoints

Health Check

GET /health

Gmail Auth Status

GET /auth/status

Fetch Gmail Emails

GET /api/gmail/emails

Analyze Emails

POST /api/ai/analyze-emails

Generate Reply

POST /api/ai/generate-reply

Create Gmail Draft

POST /api/gmail/create-draft

Get AI Analysis History

GET /api/ai/history

Delete History Item

DELETE /api/ai/history/:id

🧠 What I Learned

While building MailPilot AI, I worked on:

Google OAuth authentication
Gmail API integration
MongoDB Atlas connection and schema design
Full-stack API communication
AI-powered email workflow logic
Dashboard UI/UX design
Persistent user session handling
Secure environment variable management
Building a practical AI automation product

📸 Screenshots

Add screenshots here after uploading images to the repository.

![Landing Page](./screenshots/landing-page.png)
![Dashboard](./screenshots/dashboard.png)
![Smart Inbox](./screenshots/smart-inbox.png)
![AI Analysis](./screenshots/ai-analysis.png)
![Reply Drafts](./screenshots/reply-drafts.png)
![System Health](./screenshots/settings.png)

🧩 Future Improvements

Add OpenAI or Gemini integration for advanced LLM-based responses
Add Gmail label automation
Add follow-up reminder system
Add n8n workflow integration
Add email sentiment detection
Add deployment on Render/Vercel
Add user-specific dashboard analytics
Add role-based authentication

📌 Project Status

MailPilot AI is currently a working full-stack MVP with:

Gmail OAuth
Gmail API integration
MongoDB persistence
AI fallback analysis
Smart reply generation
Gmail draft creation
Dashboard analytics
History management
Premium UI/UX

👨‍💻 Author

Prabhuji Mishra B.Tech CSE Student Focused on AI Agents, Automation Workflows, and Full-Stack AI Projects

GitHub: javcod

⭐ Support

If you like this project, consider giving it a star on GitHub.

Latest commit

History

Folders and files

Repository files navigation