Smartgement - Smart Product Management System

Sistem manajemen produk pintar untuk UMKM dengan AI chatbot, automasi, dan analisis risiko.

Features

• AI Chatbot: Kelola produk dengan natural language (Bahasa Indonesia)
• Product CRUD: Tambah, edit, hapus, lihat produk via chat
• Automation: Operasi bulk pada produk (kosongkan stok, hapus batch, dll)
• Risk Analysis: Deteksi produk berisiko tinggi, expired, low stock
• Conversation Memory: Chatbot mengingat 7 pesan terakhir untuk konteks
• Authentication: Protected routes, user management
• Profile Management: Edit profil dan ubah password

Project Screenshots

Landing Page

Application Features

Product Management

Transaction & POS

AI Chatbot & Automation

Project Structure

smartgement/
├── backend/           # Go (Fiber v2) - API Gateway & Auth
├── aiservices/        # Python (FastAPI) - AI & Chatbot
├── frontend/          # Nuxt.js - User Interface
└── README.md

Technology Stack

Backend (Go)

• Framework: Fiber v2
• ORM: GORM
• Database: MySQL
• Auth: JWT

AI Services (Python)

• Framework: FastAPI
• ORM: SQLAlchemy
• LLM: OpenAI GPT-4o-mini
• Database: MySQL

Frontend (Nuxt.js)

• Framework: Nuxt 4
• UI: TailwindCSS + Custom CSS
• State: Composables (useAuth, useApi)

Prerequisites

• Go 1.21+
• Python 3.10+
• Node.js 18+
• MySQL 8.0+

Installation

1. Database Setup

CREATE DATABASE smartgement;

2. Backend (Go)

cd backend
go mod download
cp .env.example .env  # Configure database
go run cmd/main.go

Port: 8080

3. AI Services (Python)

cd aiservices
python -m venv .venv
.venv\Scripts\activate  # Windows
source .venv/bin/activate  # Linux/Mac

pip install -r requirements.txt

# Configure environment
cp .env.example .env
# Add your OPENAI_API_KEY

python run_server.py

Port: 8000

4. Frontend (Nuxt)

cd smartgement-fe
npm install
npm run dev

Port: 3000

Environment Variables

Backend (.env)

DB_HOST=localhost
DB_PORT=3306
DB_USER=root
DB_PASSWORD=
DB_NAME=smartgement
JWT_SECRET=your-secret-key

AI Services (.env)

DATABASE_URL=mysql+pymysql://root:@127.0.0.1:3306/smartgement
OPENAI_API_KEY=your-openai-api-key
OPENAI_MODEL=gpt-4o-mini

Frontend (nuxt.config.ts)

runtimeConfig: {
  public: {
    apiBase: 'http://localhost:8080/api',
    aiBase: 'http://localhost:8000',
  }
}

Usage

Chat Commands (Bahasa Indonesia)

Product CRUD:

- "Tampilkan semua produk saya"
- "Tambahkan produk Roti Tawar harga 15000 stok 50"
- "Ubah harga Roti Tawar jadi 12000"
- "Hapus produk Roti Tawar"

Automation:

- "Kosongkan semua produk yang mengandung tepung"
- "Hapus semua produk expired"
- "Update stok semua roti jadi 100"

Analysis:

- "Produk apa yang berisiko tinggi?"
- "Tampilkan produk yang hampir expired"

API Endpoints

Go Backend (8080)

POST   /api/auth/login
POST   /api/auth/register
GET    /api/products
POST   /api/products
PUT    /api/products/:id
DELETE /api/products/:id
GET    /api/users/:id
PUT    /api/users/:id
PUT    /api/users/:id/password

AI Services (8000)

POST   /chatbot/message
POST   /chatbot/automation/preview
POST   /chatbot/automation/execute
GET    /chatbot/chat/history
GET    /risk/assess/:product_id
GET    /risk/high-risk

Architecture

Data Flow

User (Frontend)
  ↓
Go Backend (Auth, Products)
  ↓
Python AI (Chatbot, Analysis)
  ↓
MySQL Database

Authentication Flow

1. User login → Go Backend
2. JWT token generated
3. Token stored in cookie
4. Protected routes check auth middleware
5. API calls include JWT in header

Chat Flow

1. User sends message → Frontend
2. Frontend sends last 7 messages as context
3. AI Services classify intent
4. Route to appropriate handler
5. Database operations (if needed)
6. Return response with suggested actions
7. Frontend displays messages

Key Features Explained

1. Conversation Memory

Chatbot ingat 7 pesan terakhir:

conversationHistory = messages.slice(-8, -1)

Benefit: Follow-up questions work naturally.

2. Smart Automation

Preview before execute:

1. User: "Kosongkan produk tepung"
2. AI: Shows preview (2 products affected)
3. User clicks "Eksekusi"
4. API executes automation
5. Shows success (2 products updated)

Testing

Manual Testing

Chat Interface:

# Start all services
npm run dev  # Frontend
go run cmd/main.go  # Backend
python run_server.py  # AI Services

# Test at http://localhost:3000/chat

API Testing:

# List products
$body = @{ merchant_id = "1"; message = "Tampilkan produk" } | ConvertTo-Json
Invoke-RestMethod -Uri "http://localhost:8000/chatbot/message" -Method POST -Body $body -ContentType "application/json"

Database Schema

Products

CREATE TABLE products (
  id INT AUTO_INCREMENT PRIMARY KEY,
  merchant_id INT NOT NULL,
  name VARCHAR(255) NOT NULL,
  description TEXT,
  stock INT DEFAULT 0,
  price DECIMAL(10,2),
  ingredients TEXT,
  category VARCHAR(100),
  expiration_date DATETIME,
  created_at DATETIME DEFAULT CURRENT_TIMESTAMP,
  updated_at DATETIME DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP
);

Chat History

CREATE TABLE chat_history (
  id INT AUTO_INCREMENT PRIMARY KEY,
  merchant_id VARCHAR(50) NOT NULL,
  user_message TEXT NOT NULL,
  ai_response TEXT,
  intent VARCHAR(50),
  created_at DATETIME DEFAULT CURRENT_TIMESTAMP
);

Troubleshooting

Port Already in Use

# Windows
netstat -ano | findstr :8000
taskkill /PID <PID> /F

# Linux/Mac
lsof -ti:8000 | xargs kill -9

Database Connection Error

# Check MySQL is running
mysql -u root -p

# Verify database exists
SHOW DATABASES;

OpenAI API Error

# Verify API key in .env
echo $OPENAI_API_KEY

# Test API key
curl https://api.openai.com/v1/models \
  -H "Authorization: Bearer $OPENAI_API_KEY"

Deployment

Production Checklist

• [✓] Set production environment variables

• [✓] Use secure JWT secret

• [✓] Enable HTTPS

• [✓] Configure CORS properly

• [✓] Set up database backups

• [✓] Monitor API usage (OpenAI)

• [✓] Set up error logging

• [✓] Configure rate limiting

• OpenAI for GPT-4 API

• Nuxt.js team

• FastAPI team

• Fiber team

Support

For support, email synchronizeteams88@gmail.com or open an issue.

---

Built with ❤️ for Indonesian UMKM