Passing variables from s3 buckets to reusable roles

Doesn’t seem like this is currently possible with Ansible playbooks, but it would be great if there were an include directive that would pull variables from a YAML file in s3. We have some reusable roles that some of our app stacks depend on, like a postgres RDS role. Currently, we’re defining a vault file that gets included via ‘include_vars’ in a top-level playbook and the dependent role is passed variables from there. One advantage of being able to pull from s3 is that we wouldn’t have to modify source code - source code that is vaulted, so the resultant PRs can’t be easily verified. Am I missing some mechanism to define variables from s3 (which I’ve done inside of playbooks using the s3 module followed by the include_vars module) before including a role? Is this something that might be addressed in v2?

Thanks!

Jaime