📚 Document Intelligence & Knowledge Search Hub

A production-ready MERN stack application that enables users to upload documents (PDF/TXT), extract text content, and ask AI-powered questions with answers strictly grounded in their uploaded documents.

✨ Features

🔐 Authentication

Secure Signup & Login with email and password
JWT-based authentication with token expiration
Password hashing using bcrypt
Protected routes - users can only access their own data

📄 Document Management

Upload PDF and TXT files (max 10MB)
Automatic text extraction from documents
Real-time processing status (processing, completed, failed)
Document metadata tracking (name, size, upload date)
Delete documents with file cleanup
Drag-and-drop upload interface

🤖 AI-Powered Q&A

Ask questions about uploaded documents
AI answers strictly from document content - no hallucinations
Source references for every answer
Chat-style interface with conversation history
Automatic context building from all user documents
Fallback responses when information is not found

💬 Query History

View past questions and answers
Pagination support for large histories
Delete individual queries or clear all history
Timestamp tracking for all interactions

🎨 Modern UI/UX

Responsive design - works on all devices
Clean, intuitive interface with smooth animations
Real-time updates for document processing
Loading states and error handling
Premium gradient designs and glassmorphism effects

🛠️ Tech Stack

Backend

Node.js - Runtime environment
Express.js - Web framework
MongoDB - Database
Mongoose - ODM
JWT - Authentication
bcryptjs - Password hashing
Multer - File uploads
pdf-parse - PDF text extraction
Google Gemini AI - Answer generation

Frontend

React 18 - UI library
Vite - Build tool
React Router - Navigation
Axios - HTTP client
Context API - State management
CSS3 - Styling with custom design system

📁 Project Structure

DocAssign/
├── backend/
│   ├── controllers/
│   │   ├── authController.js
│   │   ├── documentController.js
│   │   └── queryController.js
│   ├── middleware/
│   │   └── auth.js
│   ├── models/
│   │   ├── User.js
│   │   ├── Document.js
│   │   └── QueryHistory.js
│   ├── routes/
│   │   ├── auth.js
│   │   ├── document.js
│   │   └── query.js
│   ├── utils/
│   │   ├── upload.js
│   │   ├── textExtractor.js
│   │   └── aiService.js
│   ├── uploads/
│   ├── server.js
│   ├── .env
│   ├── .env.example
│   ├── .gitignore
│   └── package.json
│
└── frontend/
    ├── src/
    │   ├── components/
    │   │   ├── ChatInput.jsx
    │   │   ├── ChatMessage.jsx
    │   │   ├── DocumentList.jsx
    │   │   ├── DocumentUpload.jsx
    │   │   ├── Navigation.jsx
    │   │   └── PrivateRoute.jsx
    │   ├── context/
    │   │   └── AuthContext.jsx
    │   ├── pages/
    │   │   ├── Home.jsx
    │   │   ├── Login.jsx
    │   │   ├── Register.jsx
    │   │   ├── Dashboard.jsx
    │   │   └── Chat.jsx
    │   ├── services/
    │   │   └── api.js
    │   ├── styles/
    │   │   ├── App.css
    │   │   ├── Auth.css
    │   │   ├── Dashboard.css
    │   │   ├── Chat.css
    │   │   └── Home.css
    │   ├── App.jsx
    │   └── main.jsx
    ├── .env
    ├── index.html
    ├── package.json
    └── vite.config.js

🚀 Installation & Setup

Prerequisites

Node.js (v18 or higher)
MongoDB (local or Atlas)
Google Gemini API Key (or OpenAI API Key)

1. Clone the Repository

git clone <repository-url>
cd DocAssign

2. Backend Setup

cd backend

# Install dependencies
npm install

# Create .env file
cp .env.example .env

# Edit .env file with your configuration
# Required variables:
# - MONGO_URI: Your MongoDB connection string
# - JWT_SECRET: A strong random string
# - GEMINI_API_KEY: Your Google Gemini API key

Backend Environment Variables

Create a .env file in the backend directory:

# Server Configuration
PORT=5000

# Database
MONGO_URI=mongodb://localhost:27017/docassign
# For MongoDB Atlas: mongodb+srv://username:password@cluster.mongodb.net/docassign

# JWT Secret (Use a strong random string in production)
JWT_SECRET=your_super_secret_jwt_key_change_this_in_production

# AI Configuration
GEMINI_API_KEY=your_gemini_api_key_here
# Alternative: OPENAI_API_KEY=your_openai_api_key_here

# File Upload Configuration
MAX_FILE_SIZE=10485760

Start Backend Server

# Development mode with auto-reload
npm run dev

# Production mode
npm start

The backend server will run on http://localhost:5000

3. Frontend Setup

cd frontend

# Install dependencies
npm install

# Create .env file
echo "VITE_API_URL=http://localhost:5000" > .env

# Start development server
npm run dev

The frontend will run on http://localhost:5173

🔑 Getting API Keys

Google Gemini API Key

Visit Google AI Studio
Sign in with your Google account
Click "Create API Key"
Copy the key and add it to your .env file

Alternative: OpenAI API Key

Visit OpenAI Platform
Sign in or create an account
Create a new API key
Copy the key and add it to your .env file
Uncomment the OpenAI code in backend/utils/aiService.js

📖 How It Works

Document Processing Flow

Upload: User uploads a PDF or TXT file
Storage: File is saved to the uploads/ directory
Database Entry: Document metadata is stored in MongoDB with status "processing"
Text Extraction:
- PDF files: Extracted using pdf-parse library
- TXT files: Read directly from filesystem
Update: Document status updated to "completed" or "failed"
Ready: Document text is now available for AI queries

AI Question Answering Flow

User Question: User asks a question in the chat interface
Context Building: System retrieves all completed documents for the user
Prompt Construction: Creates a prompt with:
- System instructions (answer only from documents)
- All document texts with metadata
- User's question
AI Generation: Sends to Google Gemini (or OpenAI)
Reference Extraction: Identifies relevant document excerpts
Response: Returns answer with source references
History: Saves question, answer, and references to database

AI Limitations & Constraints

✅ What the AI Does

Answers questions strictly based on uploaded document content
Provides source references for every answer
Indicates when information is not found in documents

❌ What the AI Does NOT Do

Generate answers from general knowledge
Make assumptions beyond document content
Process scanned PDFs (OCR not implemented)
Understand images or charts in PDFs

🔍 Known Constraints

No OCR: Scanned PDFs without text layer won't be processed
Basic Matching: Uses simple keyword matching for references (no vector DB)
Context Limit: Very large documents may exceed AI context window
English-focused: Best results with English text
No Multi-modal: Cannot process images, tables, or charts

🎯 Usage Guide

1. Create an Account

Navigate to the home page
Click "Get Started" or "Sign Up"
Enter your email and password (min 6 characters)
Click "Create Account"

2. Upload Documents

Go to the Dashboard
Drag and drop a PDF or TXT file, or click "Choose File"
Click "Upload Document"
Wait for processing to complete (status will update automatically)

3. Ask Questions

Navigate to the Chat page
Type your question in the input field
Press Enter or click the send button
View the AI's answer with source references

4. Manage Documents

View all documents in the Dashboard
Check processing status
Delete documents you no longer need

5. View History

All questions and answers are saved automatically
Scroll through chat history
Clear history if needed

🔒 Security Features

Password Hashing: All passwords hashed with bcrypt (10 salt rounds)
JWT Authentication: Secure token-based auth with 7-day expiration
Protected Routes: Backend validates JWT on all protected endpoints
User Isolation: Users can only access their own documents and queries
File Validation: Only PDF and TXT files allowed, max 10MB
Error Handling: Comprehensive error handling without exposing sensitive data

🧪 Testing the Application

Test Document Upload

Create a simple TXT file with some content
Upload it via the Dashboard
Wait for "Completed" status

Test AI Q&A

Upload a document with known content
Ask a specific question about that content
Verify the answer matches the document
Check that references are provided

Test AI Limitations

Ask a question about content NOT in your documents
Verify you get: "The uploaded documents do not contain enough information to answer this."

🐛 Troubleshooting

Backend won't start

Check MongoDB is running
Verify .env file exists and has correct values
Ensure port 5000 is not in use

Frontend can't connect to backend

Verify backend is running on port 5000
Check VITE_API_URL in frontend .env
Check browser console for CORS errors

Document processing fails

Check file is valid PDF or TXT
Verify file size is under 10MB
Check backend logs for errors

AI answers are generic

Verify your API key is correct
Check document was processed successfully
Review backend logs for AI service errors

📝 API Documentation

Authentication Endpoints

POST `/api/auth/register`

Register a new user

{
  "email": "user@example.com",
  "password": "password123"
}

POST `/api/auth/login`

Login user

{
  "email": "user@example.com",
  "password": "password123"
}

GET `/api/auth/profile`

Get current user profile (requires auth)

Document Endpoints

POST `/api/documents/upload`

Upload a document (requires auth, multipart/form-data)

GET `/api/documents`

Get all user documents (requires auth)

GET `/api/documents/:id`

Get single document (requires auth)

DELETE `/api/documents/:id`

Delete document (requires auth)

Query Endpoints

POST `/api/queries/ask`

Ask a question (requires auth)

{
  "question": "What is the main topic?"
}

GET `/api/queries`

Get query history (requires auth)

DELETE `/api/queries/:id`

Delete single query (requires auth)

DELETE `/api/queries/clear`

Clear all history (requires auth)

🚀 Deployment

Backend Deployment (Render/Railway/Heroku)

Push code to GitHub
Connect repository to hosting platform
Set environment variables
Deploy

Frontend Deployment (Vercel/Netlify)

Push code to GitHub
Connect repository to hosting platform
Set build command: npm run build
Set output directory: dist
Add environment variable: VITE_API_URL=<your-backend-url>
Deploy

MongoDB Atlas Setup

Create account at MongoDB Atlas
Create a cluster
Create database user
Whitelist IP addresses (or allow all for development)
Get connection string
Update MONGO_URI in backend .env

🤝 Contributing

Contributions are welcome! Please follow these steps:

Fork the repository
Create a feature branch
Commit your changes
Push to the branch
Open a Pull Request

📄 License

This project is licensed under the MIT License.

👨‍💻 Author

Built with ❤️ using the MERN stack

🙏 Acknowledgments

Google Gemini AI for intelligent answer generation
pdf-parse library for PDF text extraction
MongoDB for flexible document storage
React and Vite for modern frontend development

Note: This is a demonstration project. For production use, consider adding:

Rate limiting
File virus scanning
Vector database for better semantic search
OCR for scanned PDFs
Multi-language support
User email verification
Password reset functionality
Document sharing features
Advanced analytics

Name		Name	Last commit message	Last commit date
backend		backend
frontend		frontend
.gitignore		.gitignore
.vercelignore		.vercelignore
DEPLOYMENT.md		DEPLOYMENT.md
ENV_SETUP.md		ENV_SETUP.md
README.md		README.md
render.yaml		render.yaml
vercel.json		vercel.json

Latest commit

History

Folders and files

Repository files navigation