Yea I don't think it will work.

When you execute a shell command, you are working on the OS level. Linux does not know about DBFS or Azure or AWS or ...

If you want to do this, you have to mount the data lake in Linux, and this is not that easy.

Databricks (or the Spark application) bootstraps all this and gives you dbfs so you do not have to worry about connectivity.

May I ask why you want to do this with a shell command?