> ## Documentation Index
> Fetch the complete documentation index at: https://docs.trunk.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Flaky Tests: Detect Consistently Failing Tests as Broken

> Threshold monitors can now classify consistently failing tests as Broken, distinct from Flaky.

**Tests that consistently fail can now be classified as Broken, distinct from Flaky.** This gives teams a clearer picture of test health when they need to triage real regressions separately from intermittent flakes.

When creating or editing a threshold monitor, set the **Detection type** to **Broken**. Monitors configured this way mark tests as **Broken** (shown in red) once the failure threshold is exceeded. Each monitor card shows its detection type with a color-coded badge: blue for Flaky, red for Broken.

Test status priority is **Broken > Flaky > Healthy**, so a test flagged by both a broken and a flaky monitor shows as Broken until the broken monitor resolves. The `v2.test_case.status_changed` webhook now includes `BROKEN` as a status value, so you can route broken tests through a separate triage workflow than flaky ones.

[Read the docs](/flaky-tests/detection/failure-rate-monitor#action) to learn more.