This argument isn't very compelling. Shell scripts are great at tricking you into believing you've got something 100% working when in fact you've only accounted for the happy path and maybe two or three failure cases. This is the primary reason shell scripts are so often broken. There are tons of assumptions baked in which are easily violated.

If your shell script keeps breaking, and you keep fixing it by writing more shell, then you're in an abusive relationship with your programming environment.