Pick your CI. This takes minutes, not a sprint.
One integration. No pipeline rewrite, no sales call.
GitHub Actions and GitLab CI are live.
Pick a provider to continue
Authorize FlakeKill to watch your runs.
We read your test-run history to find the liars. You can revoke access anytime. The exact scope of what we read is being finalized.
No credit card. No email required to start.
Reading your runs. Hunting for liars.
We’re replaying your test history to spot the tests that fail on unchanged code.
Run-cell grid filling in progressively as FlakeKill scans recent CI runs for flaky patterns.
Scanned 40 of 40 runs
Large history? This can take a couple of minutes. Still working.
Found the liars. Here’s your report.
These tests failed on unchanged code. Quarantine kicks in within one run of detection. The worst offender is flagged to fix first.
A first flake report in the FlakeKill dashboard: five CI tests ranked by flakiness — two flaky tests queued for quarantine, one at 40% and flagged fix first, one at 35% — and three stable at 0%, each with a run-history sparkline.
% of runs that failed on unchanged code
2 flaky · 2 queued · quarantine within one run of detection
No liars found. Your red already means broken.
We scanned your runs and every failure lined up with an actual code change. Nothing to quarantine — which is the good kind of boring.
Flakes creep in over time. When one does, it’s quarantined within one run of detection.
Couldn’t connect. Not your fault.
Authorization didn’t go through — we couldn’t reach your CI’s run history.
If it keeps failing, your CI may have revoked access mid-request.
← Back to provider list