For non-reproducible flaky tests, Findit does not take any auto-actions such as updating bugs with findings. In this case, it is unclear if Findit bailed out, had an error, or had no findings, etc. such that a user observing the bug queue on SoM may not know what to expect or whether to expect anything at all.

For flaky tests that are not reproducible, Findit should have a confidence score too (e.g. ran with 1000 iterations and still all 100% passing), then this should be surfaced as well so the sheriff can easily gauge the severity of the flakiness