I do not think that I have enough patience to write an example. I will try to explain the algorithm in a simplified way, as I understood it.
kaboonus correctly indicated the code from which I would start the investigation.
Thus, we have an MotionParam object with data: start and end element + mouse button.
How does this data turn into action?
Omit the travel details of our object motion. As a result, he gets here:
var BeginTime = GetTimeStopwatch();
var Result = Motor.ActSequenceMotion(motion.AsSequenceMotion(MemoryMeasurementLast?.Value?.MemoryMeasurement));
var EndTime = GetTimeStopwatch();
First, our MotionParam object (motion) calls AsSequenceMotion().
As far as I understand, this function converts MP to the list of more simple Motion objects:
After this bot call ActSequenceMotion() function, which uses WindowsInput.InputSimulator() and converts Motion list to real keystrokes / mouse clicks.
It looks simple, if you do not go into details. 