Turn ChatGPT Integration

Description:

This application serves as an integration layer between the Turn UI interface and any of OpenAI's chat completion models (defaulting to GPT-3.5-turbo). The main functionality is to transform incoming messages from Turn UI and get the model's suggested responses which are then returned as structured replies.

Setup:

1. Environment Variables: Ensure you have the following environment variables set:

   • OPENAI_API_KEY: Your OpenAI API Key
   • SYSTEM_PROMPT: The initial system prompt for the ChatGPT model (default is a health-oriented prompt).
   • MODEL: The OpenAI model being used (default is "gpt-3.5-turbo").
   • NUMBER_OF_MESSAGES_FOR_CONTEXT: Number of most recent inbound and outbound messages considered for context (default is 4, maximum value is 10).
   • CHATGPT_TIMEOUT: Timeout for OpenAI API calls in seconds (default is 10 seconds).
   • PORT: The port on which the application runs (default is 8080).

2. Install necessary packages:

   python -m pip install -r requirements.txt
   

3. Run the application:

   python main.py
   

The application will now be running, awaiting incoming HTTP requests on the specified port.

Endpoints:

1. /: A simple endpoint that informs users about the primary endpoint for integration.

2. /integration (POST): The main integration point where incoming messages from Turn UI are transformed and fed to the ChatGPT model for suggested responses.

Note:

Always keep your OPENAI_API_KEY confidential. Avoid hardcoding or exposing it in any public repositories.