Built with Assistance from

ChatGPT

Public Chat Link

GPT Custom Server

This project sets up a server to interact with a custom GPT instance, allowing it to read from and write to files, and run Jest tests programmatically.

Prerequisites

• Docker
• ngrok account

Getting Started

1. Clone the repository:

git clone <repository-url>
cd project

2. Create a .env file with your ngrok token and subdomain (if applicable):

echo "NGROK_AUTHTOKEN=your_ngrok_authtoken" > .env
echo "NGROK_SUBDOMAIN=your_subdomain" >> .env

3. Build and start the Docker containers using Docker Compose:

docker-compose up --build

4. The server should now be running on http://localhost:3000 and exposed via ngrok. If you set a subdomain, the public URL will be https://your_subdomain.ngrok.io. Otherwise, check the ngrok dashboard or logs to get the public URL.

Ngrok Configuration

The ngrok service is configured in the docker-compose.yaml file. Here is the relevant configuration:

ngrok:
  image: ngrok/ngrok:latest
  command: http server:3000 --authtoken ${NGROK_AUTH} --subdomain ${NGROK_DOMAIN}
  restart: unless-stopped
  ports:
    - "4040:4040"
  depends_on:
    - server

Environment Variables:

• NGROK_AUTH: Your ngrok authentication token.
• NGROK_DOMAIN: Your desired subdomain for the ngrok tunnel.

Usage:

1. Ensure NGROK_AUTH and NGROK_DOMAIN are set in your environment.
2. Run ./start.sh /path/to/project to start the services, including ngrok.

The ngrok tunnel will forward traffic to the server service running on port 3000 and expose the ngrok web interface on port 4040.