🤖 Local Chat API

A lightweight Flask-based REST API that wraps OpenAI's GPT-4o-mini model and exposes a simple HTTP endpoint for chat interactions.

---

📁 Project Structure

.
├── local_chat_api.py   # Main Flask application
├── requirements.txt    # Python dependencies
└── start.sh            # Production server startup script

---

⚙️ Prerequisites

• Python 3.8+
• An OpenAI API key

---

🚀 Setup & Installation

1. Clone the repository

git clone <your-repo-url>
cd <your-repo-folder>

2. Install dependencies

pip install -r requirements.txt

3. Set your OpenAI API key

export OPENAI_API_KEY="sk-..."

Tip: Add this line to your .bashrc or .env file to persist it across sessions.

---

▶️ Running the Server

Development mode

python local_chat_api.py

Server will start at http://0.0.0.0:5000

Production mode (Gunicorn)

bash start.sh

This launches the app with 2 Gunicorn workers on port 5000.

---

📡 API Reference

GET /chat

Send a message to GPT-4o-mini and receive a reply.

Query Parameters

Parameter	Type	Required	Description
text	string	Yes	The message to send

Example Request

curl "http://localhost:5000/chat?text=Hello%2C+how+are+you%3F"

Example Response

{
  "reply": "I'm doing well, thank you! How can I help you today?"
}

---

🔧 Configuration

Variable	Description	Default
OPENAI_API_KEY	Your OpenAI API key (required)	—
PORT	Server port (set in start.sh)	5000
model	OpenAI model used	gpt-4o-mini

---

📦 Dependencies

Package	Version	Purpose
flask	3.0.0	Web framework
requests	2.31.0	HTTP client for OpenAI API
gunicorn	21.2.0	Production WSGI server

---

⚠️ Notes & Limitations

• The /chat endpoint uses a GET request with query parameters. For longer messages, consider switching to a POST endpoint with a JSON body to avoid URL length limits.
• There is no authentication on the /chat endpoint — do not expose this server publicly without adding proper auth middleware.
• Error handling is minimal; if the OpenAI API returns an unexpected response, the server will raise an unhandled exception.

---

🛡️ Security Recommendations

• Never hardcode your OPENAI_API_KEY in source code.
• Use environment variables or a secrets manager (e.g., AWS Secrets Manager, .env with python-dotenv).
• Add rate limiting (e.g., Flask-Limiter) before deploying to production.

---

📄 License

MIT — feel free to use and modify.