Code health for google-research/harness2
Repowise scores google-research/harness2 at 7.9 out of 10 for code health, which reads as good, alongside maintainability 7.0 and static performance 8.9 on the same scale. There are 254 open findings across the three pillars, and the lowest-scoring file is schedule.py at 5.5. Scores are computed from complexity, duplication, churn and git fix history, not from a linter.