View as Markdown

CI Insights

Optimize your CI run time, catch flaky tests, and give developers actionable insights so your team ships code faster.


CI Insights helps you monitor your pipelines, detect flaky tests and jobs, and improve your CI efficiency. This guide walks you through enabling CI Insights via GitHub and covers basic configuration steps.

First time here? Follow these 3 steps to see data fast:

  1. Install the Mergify GitHub App on a repo with existing CI.
  2. Ensure your tests produce a supported format (JUnit XML, pytest, Jest, etc.); pick a framework below if needed.
  3. Re-run a workflow (or push a trivial commit), then open CI Insights in the dashboard to view your jobs and tests.

CI Insights flags a flaky job when the same job runs more than once on the same commit (SHA1) and the runs end with different conclusions: one fails while another succeeds, even though the code did not change. Cancelled runs are ignored, since a cancellation says nothing about the code.

Tests use the same rule, with one addition: a flaky test has to have failed on some runs of a commit and passed on others. A test that is skipped in one run and passes in another stays healthy.

To set up systematic flaky test detection in your CI pipeline, see our dedicated Flaky Test Detection guide.

Select your CI system below to enable CI Insights and configure flaky test detection.

To capture test-level data from your CI pipeline, configure your test framework to upload results to Mergify. See the Test Insights test framework guides for framework-specific setup instructions.

Was this page helpful?