Is there any ideas for creating an interface for stack outputs? I caught myself using the automation API to orchestrate 100+ micro-stacks and weaving outputs together and staging dependent resources, but it seems like I am really just creating a layer on top of what Pulumi already does pretty well with outputs.
My use case is solving the orchestration issue that arises when moving from a mono-repo to micro-stacks. From one side, we want to make workflows efficient for dev teams, but it comes at a cost to us as the operators as we need to pick up the orchestration "slack". Stuff like provisioning dev environments can be tricky if we need to trigger 60+ pipelines in other repositories. If dev teams services depend on other teams services it becomes even more opaque what to do.
An interface for stack outputs may not be the only thing i would need to solve this. I would also need a way to trigger stacks in sequence which is what i use the automation API for right now.
I guess what i need is some automation at a level higher than automation API proposes.
If a pulumi-person has 10 minutes some time, i would love to explain what i am talking about on a video call.