MIT License. Do not use this abusively and be respectful of public servers acceptable usage policies.
Based upon https://github.com/knrd1/chatgpt/tree/master with changes made by Chatgpt-4.
While the bot works, sometimes connecting is flaky, at which point the process needs to be killed and retried. It almost always works within 3 tries if it does not work at first.
Multiple bots have been tested in one room. I have seen them make a comment about server load and then they talk to each other on how to improve server performance and then plan a trip to Rome.
I plan on adding Kubernetes deployment functionality to the bot because who wouldn't love having N number of assistant chatbots that can read server details running on your cluster?