Hey
This sounds to me like expected behaviour. What I mean by that is the way the slot swap works is if you’ve deployed the code to your staging slot and swap this to production, the staging slot will now be whatever was last on that production slot in terms of code. If the last code was just the default app service code then that’s what will be on the swap.
You need to deploy your code again to the slot so both have matching code. That way, when you deploy changes next time and swap you will have a revert method to the old working state in addition to the zero downtime deployment benefit.
You can read more here: https://learn.microsoft.com/en-us/azure/app-service/deploy-staging-slots?tabs=portal#what-happens-during-a-swap