Custom AI generation command

• iTerm2 version: latest
• OS version: Mac OS 15.5

Could it be possible to extend the AI plugin to allow specifying an arbitrary command to run to get AI output?

Right now we can only point to OpenAI compatible endpoints which works fine though I was kind of wondering if I could plumb it through something like: https://docs.github.com/en/copilot/using-github-copilot/using-github-copilot-in-the-command-line

I have access to copilot via the gh cli, so in theory if I could tell iterm2's AI plugin to call that cli (with possibly predefined boiler text) to get output.

This type of functionality would in theory also make it easier to use any other AI endpoint/cli/etc since we would set the cli/script to call to get AI responses.

I don't consider this a replacement for OpenAI integration, but rather an alternative.

As an alternative I could just call the cli I want directly, but the UI for AI via iterm2 was pretty nice so I figured I'd ask.

Thanks for considering.