Chat AI - Professional Python and Oracle SQL Assistant

Full-stack AI chat application for Python developers and database administrators. It provides a ChatGPT-style UI, secure authentication, and routing between n8n, Groq, and Ollama.

• Modern, responsive UI built with React 19 and Tailwind CSS.
• Provider routing between n8n workflows, Groq Cloud, and local Ollama.
• JWT authentication with bcrypt password hashing and rate limiting.
• Syntax-highlighted code blocks with one-click copy.
• Pre-configured for Python (cx_Oracle) and Oracle SQL optimization.
• Usage limits (10 chats per user, 20 messages per chat) to control local resources.

• Frontend: React 19, Vite, Tailwind CSS, React Markdown.
• Backend: Node.js (ESM), Express, MongoDB.
• Security: Helmet, CORS, Express Rate Limit, JWT.
• AI integration: n8n Webhook, Groq API, Ollama API.

Chat ai/
|-- backend/                  # Express API and AI routing
|-- frontend/                 # React application (Vite)
|-- n8n-chat-ai-template.json # Ready-to-import n8n workflow
`-- README.md                 # Documentation

1. Navigate to the backend directory:

   cd backend

2. Configure environment variables:

   cp .env.example .env

   Ensure JWT_SECRET is at least 32 characters long.
3. Install dependencies:

   npm install

4. Start the server:

   npm run dev

   API runs at http://localhost:4000.

1. Navigate to the frontend directory:

   cd frontend

2. Configure API URL:

   cp .env.example .env

   Set VITE_API_URL=http://localhost:4000/api.
3. Install dependencies:

   npm install

4. Run in development mode:

   npm run dev

   Open http://localhost:5173 in your browser.

The backend sends a POST request to N8N_WEBHOOK_URL with chat history and message.

• Import: use n8n-chat-ai-template.json in n8n.
• Response format: n8n must return { "reply": "..." }.

High-speed inference using Llama 3 models on Groq Cloud.

• Set GROQ_API_KEY in backend/.env.

Local LLM execution.

• Ensure Ollama is running (ollama serve).
• Set OLLAMA_URL (usually http://localhost:11434/api/chat).

• Security headers via Helmet and aggressive rate limiting on auth routes.
• Backend validates message lengths and JSON payloads.
• MongoDB connection includes retry logic and index setup.
• Structured logging for debugging and monitoring.

Distributed under the MIT License. See LICENSE for details (if available).