I’d use an action.
All of the logic takes place either from a trigger or from context of when it is pressed.
An axis is more for continuous and/or smooth input like a joystick.
To get started just try to get your logic in place by using print strings only. You can have the blueprint open and set to debug the actor while in PIE. This way you can easily see the execution wires firing. It’s good to plug a print string into every execution pin even if you are not using them of the input action. That way you can see exactly how it works in real time.