* Create a script to run the tests in a loop and capture logs + output on failure * Apply PR feedback