My current understanding as per my conversation with @karencfv last week is that system will be the root command in the CLI which these operator commands will be grouped under. As such, I've prepended system to the method name.

I think we should not prepend system_ to all these methods, but I don't feel that strongly. I also think we should put them all under a system tag (we have too many tags right now, again IMO).

With regard to the CLI, our intention is to generate a big chunk of it... and to manually intervene. For example, we might take the clap hierarchy and then hoist all the subtrees with the system tag under a system subcommand.

All this is to say: we have several different consumers for the operationId: the CLI, the API, and the SDKs. I think in this case it makes more sense to do some manual intervention in the CLI than it does to impose a prefix for the API and SDKs, but I'd love to have that discussion.

Sounds good! I'll remove it for this PR and update the tags.

thanks! there's tags review on my list. I should find a way to make these pending work items visible...

I agree! system_ does not have to be prepended. Having the endpoint path start with /system/ should be enough imo :)

This will also make it clear for the terraform provider resource schema which is not (and most likely never be) generated based on tags or operation IDs.

Thanks for working on this @zephraph btw!!!!