This message was deleted.

I’ll give a simple example but I’m sure our users have much more interesting use-cases. Or customer engineering organization has a monorepo with a bunch of demo Pulumi projects in it, and we often add new projects to the repo and we want to make sure all our demos are always ready to go. So, I have an automation API based program that finds all the projects and then runs a

pulumi up

and then a

pulumi destroy

for each project and reports to our slack channel the success or failure of those runs. This is kicked off once a week by a CI/CD pipeline that is triggered by a cron trigger.

Nice idea . I am wondering if the same can be achieved using other automation tools ( bash script, CLIs, pipelines etc) .

Sure they can. And that’s generally the standard approach. There’s no fundamentally wrong answer here. But it’s nice to have the options that the automation api sdk provides. And that I can have both my stack code and orchestration code in a standard programming language.