Tried using the skulk sensor with an elytra. Used Wool on 2 opposing sides of the sensor. It detected elytra flights in about 50% of my tests, with a not-excluded sensor detecting all flight. A correlation with speed seems to exist.
Steps to reproduce:
1. place a skulk sensor
2. place wool on the north and south / west and east side of the sensor
3. equip an elytra
4. fly past the sensor on an uncovered side
Sensor only triggered in about 50 % of tests
Sensor triggering in all tests
demonstration of the test with comparison to a non-prepared sensor
higher speeds while flying seem to make non-detection more likely
Other Events like walking still trigger the sensor consistently
potential cause: maybe the event is triggered once when entering the radius of the sensor but not reapplied when passing the wool boundary from where it should be detectable again?
Tests with only one wool block placed in the direction from which the flight came had the sensor trigger at a later point in time than expected.