Skip to main content
You can now flag tests as soon as they accumulate a configured number of failures — no failure-rate math, no minimum sample size required. The new failure count monitor reacts to individual failures within a rolling time window, making it well-suited for stable branches like main where any single failure is meaningful. Set it up with a failure count (default: 1), a window, a resolution timeout, and the branches to watch. Common uses:
  • Stable branch alerting — flag any test that fails on main, even once.
  • Post-merge regression detection — catch regressions before a failure-rate monitor accumulates enough data to trigger.
  • Merge queue / release branches — quick alerting on branches where failures are suspicious by definition.
A live Preview panel shows which tests would be flagged under your current settings before you save, so you can tune the threshold with confidence. Read the docs to learn more.