Support Ticket System

A full-stack support ticket application with LLM-assisted ticket classification.

Quick Start (From a Fresh Clone)

git clone https://github.com/shubh100802/Support-Ticket-System.git
cd Support-Ticket-System

Create .env in project root:

LLM_API_KEY=your_api_key
LLM_MODEL=

Run with Docker:

docker compose up --build

Or follow the local setup steps in the Run Locally (Without Docker) section below.

Features

• Create support tickets
• Auto-suggest category and priority from description using an LLM
• Override suggested values before submission
• List tickets with combined filters and search
• Update ticket status
• View aggregated ticket statistics

Stack

• Backend: Django + Django REST Framework
• Database: PostgreSQL
• Frontend: React (Vite)
• LLM: OpenAI Chat Completions API
• Containers: Docker + Docker Compose

Repository Structure

backend/             Django API
frontend/            React app
docker-compose.yml   Multi-service setup

Prerequisites

• Docker Desktop (or Docker Engine + Compose plugin)
• Optional for local non-Docker run:
  • Python 3.11+
  • Node.js 20+
  • npm 10+
  • PostgreSQL 14+

Run with Docker (Recommended)

1. Build and start services:

docker compose up --build

2. Open:

• Frontend: http://localhost:5173
• Backend API: http://localhost:8000/api
• Django admin: http://localhost:8000/admin

To stop services:

docker compose down

To stop and remove DB volume:

docker compose down -v

Run Locally (Without Docker)

1. Backend

From the project root:

cd backend
python -m venv .venv

Activate virtual environment:

• macOS/Linux:

source .venv/bin/activate

• Windows PowerShell:

.venv\Scripts\Activate.ps1

Install dependencies and run migrations:

pip install -r requirements.txt
python manage.py migrate
python manage.py runserver

2. Frontend

In a new terminal:

cd frontend
npm install
npm run dev

3. Database for local mode

Backend reads PostgreSQL from environment variables in backend/.env.example. If those variables are not set, it falls back to SQLite for quick local testing.

API Endpoints

• POST /api/tickets/
• GET /api/tickets/
  • Query params: category, priority, status, search
• PATCH /api/tickets/<id>/
• GET /api/tickets/stats/
• POST /api/tickets/classify/

Classify request:

{
  "description": "Card charged twice this month"
}

Classify response:

{
  "suggested_category": "billing",
  "suggested_priority": "high"
}

Fallback response when LLM is unavailable:

{
  "suggested_category": null,
  "suggested_priority": null
}

Notes

• Stats endpoint uses database-level aggregation.
• Ticket field constraints are enforced at DB level.
• CORS is configured for local frontend development on localhost:5173.