Sometimes the tests don't run correctly either because of bugs or human mistakes. I think it'd be helpful to color the result in a different color (say, yellow?) as a warning. There was one or two cases where I ran the test, saw it was green and left it. Turned out it actually didn't run.
Are you suggesting printing the result in yellow for the entire suite if there's 0 assertions?
No, just make the line 0 tests, 0 assertions, 0 failures another color than green. Because lots of times I make a change and run it again and take green as an indication of everything going well. Last time it costed me a bit. I think whatever happens, having 0 test is an edge case with no good justification. So it should be colored differently.
0 tests, 0 assertions, 0 failures
This is also a good stop gap solution for all the past and future issues where glitches would return 0 tests done.