Currently, it's a bit hard to evolve deployed Power Automate Scripts without a larger risk involved.
Power Automate should have a way to run in Test mode that does not risk deployed scripts.
But scripts must be saved before running, so there is significant risk of breaking running apps.
I propose 2 facilities to ease out enhancements to existin Power Automate scripts:
runtime variable testMode: default false, becomes true if the user uses the "Test" command when editing. Is also writeable, so the user can override its setting.
This allows the user to wraps a series of actions in a IF block, so it does not break the existing flow when it is running outside of test mode.
It is also writeable to give the user more flexibility, to manually start the testMode only if a specific set of conditions is met. Useful for automatic trigger testing, like "when user receives an email", to trigger only if a specially crafted email is sent, but not otherwise.
Configure: Run After: Test Mode - This is a simple way to restrict a single action to only run if it is in test mode, removing the need to create an IF block and test for the variable.
Both help solving the problem in the same way: they restrict a code block to only run if it is being tested. But they have different ways around that are more convenient in different scenarios, and can coexist.