Skip to main content
The Trunk browser extension now surfaces per-test failure details and an AI investigation prompt directly on your GitHub pull request page. When Trunk detects test failures on a commit, the extension’s test-results panel gains a new Failure details section below the summary counts. Expand it to see:
  • Test name and file: exactly which test failed and where it lives in the repository.
  • Error snippet: the first lines of the failure output, enough to see the assertion or exception at a glance.
  • Prev / Next navigation: step through all failures for the commit without leaving the PR page.
  • Copy Prompt: copies an LLM-friendly investigation prompt to your clipboard. Paste it into any AI assistant (Cursor, Claude, etc.) to start a structured root-cause investigation with the failure context already pre-filled.
The failure details pull from the same test_run_v2 data that powers the Trunk Flaky Tests dashboard, so quarantined runs are excluded automatically. The extension shows a “still processing” state when test ingestion is in flight and results are not yet final. This pairs with the Test details panel, which shows aggregate pass/fail/quarantined/flaky counts per commit. Failure details drill one level further, down to individual failing tests.