Closed cfuehrmann closed 4 months ago
You are supposed to use --authCopilot
to acquire the token. But there is an oversight in its implementation. It expects you to already provide a token. However, this token is not used so you can just specify any random token and use the function to get an actual token and store that in the token env variable.
TLDR: execute HANDLER=copilot COPILOT_API_KEY=123 helix-gpt --authCopilot
and use the printed token.
EDIT: see #31
Thanks ever so much! I have it all working now. Great project!
Please forgive this stupid question, I seem to be having a blackout about the required API key. I pay for GitHub Copilot, and for a long time I've been using Copilot in Visual Studio, Visual Studio, and Neovim. In your instructions, you mention a required API key. But I can't figure out for the life of me how to obtain that key. I can find no corresponding function in my GitHub account. How do I find the key?
And is this key the same as the "token" you mention in the context of
helix-gpt --authCopilot
? Or are key and token two different things?What is supposed to happen when I use
helix-gpt --authCopilot
? What I get is "error: no handler key provided". But isn't the very command supposed to get me the key?Can you maybe clarify your instructions in this regard?