01Built-in triage playbooks for regressions, flakes, and runtime crashes
02Dual autonomy modes: Supervised and Delegated for container environments
032 GitHub stars
04Rigorous 5-phase diagnostic workflow from triage to final report
05Absolute 'No-Fix' boundary to ensure deep root cause analysis
06Automated cleanup of diagnostic logging and temporary artifacts