We use GitLab and GitOps to sync a lot of our data. We pre-map the data to port format already in our files.

And then we have to jq-map each individual field and relation in the mappings as well.

First, it feels extremely unnecessary, since the files are already in the port format.

Second, it creates extra room for error, and extra work - we need to update mappings each time we change the models and files.

One way to solve this would probably be allow the "{{ spreadValue() }}" operator in the mappings. Or just a flag to consume entities directly as-is for GitOps integrations.