Ansible when/Conditional Logic Review Prompt

You are a senior Ansible engineer who treats `when` as the most error-prone line in a task, because a conditional that quietly evaluates wrong doesn't error — it skips a task that should run or runs one that shouldn't, and the play still shows green.

I will paste tasks with conditionals. Review and rewrite them for correctness.

Steps:

1. **Type correctness**: identify comparisons that assume string vs. int vs. bool and fix them (e.g. `| int`, `| bool`, quoting), since a var from vars_prompt, extra-vars, or a register is often a string.
2. **Undefined safety**: guard against undefined variables with `is defined` / `| default(...)` so the condition fails predictably instead of erroring or evaluating surprisingly.
3. **Boolean truthiness**: replace reliance on implicit truthiness (the classic `"false"` string trap) with explicit `| bool` comparisons for anything gating a destructive action.
4. **when + loop interaction**: clarify that `when` with a loop is evaluated per item, and fix conditions that assume it gates the whole task.
5. **Registered results**: use the right keys (`rc`, `stdout`, `changed`, `failed`, `is skipped`) and handle the skipped/failed cases.
6. **Prove it**: add a temporary `debug` showing the resolved condition value for each realistic input before removing it.

Fill in:
- The tasks/conditionals to review: [PASTE]
- Where the vars come from: [extra-vars / register / inventory / prompt]
- Which tasks are destructive: [LIST]

Output format: the rewritten tasks, an explanation of each bug found, and the debug-based verification that proves the conditions resolve as intended.