Skip to content

Latest commit

 

History

History
37 lines (23 loc) · 1.11 KB

README.md

File metadata and controls

37 lines (23 loc) · 1.11 KB

Automated SMS Reply System

This system automates the process of replying to incoming SMS messages using Pushbullet and OpenAI's GPT-3.

Overview

Given the influx of messages in the medical domain, this system listens for incoming SMS via Pushbullet's WebSocket API and generates human-like responses using OpenAI's GPT-3. The response guides patients to book their appointments online.

Prerequisites

  • Python 3.x
  • openai, requests, and websockets Python libraries. Install them using:
pip install openai requests websockets

Configuration

  1. Set up environment variables:

    • OPENAI_API_KEY: Your OpenAI API key.
    • PUSHBULLET_ACCESS_TOKEN: Your Pushbullet access token.
    • DEVICE_ID: The device identifier for Pushbullet.

  2. Clone the repository and navigate to the project directory.

Usage

Run the script:

python main.py

Disclaimer

This project is a side experiment and is not intended for production use, especially in the medical field. Ensure data privacy and consider self-hosting for sensitive applications. Always prioritize the privacy and security of patient data.