v2026.10.22
TL;DR
Thunders AI can now execute your tests and debug them until they pass, you can take control of the live browser at a failed step and record the fix, and authenticated profile pools give each parallel run its own login session.
FEATURE
- Thunders AI runs and debugs your tests: Thunders AI can now run your test, find why it fails, fix the steps, and run it again until it passes. A higher effort level allows more tries.
- Take control of a failed step: When a step fails, you can now take control of the live browser and record the correct step.
- API app spec: Give an API app your API spec, and Thunders AI uses it to write API tests. Thunders also checks the base URL while you set up the app.
- Browser image releases: Full self-hosted customers now get the browser image with the weekly release. Thunders now emails your contact when a new browser image is in your registry.
- Authenticated profile pools: Group interchangeable login profiles in a pool, and each parallel run gets its own session. Apps that allow one session per account no longer force your tests to run one at a time. Profiles also work like variables now: set one per project or per environment, so one login step works on every environment. If a saved session fails, a step can continue with a warning, so your test can sign in another way.
IMPROVEMENT
- Browse on by default: Thunders AI now browses your web and mobile apps by default when it creates test cases, so its steps match what your app really shows.
- Thunders AI for API apps: Thunders AI now uses your API spec and live calls to write API tests that fit your API.
- Xray: Xray Cloud and Xray Data Center now appear as a knowledge source and as a test result sync.
- Date pickers: Thunders now works with custom date pickers in your app during test execution. It opens the picker, navigates it, and clicks the date.
- Run screenshots: You now see which element each step acted on, right on the run screenshot.
FIX
- Conditional steps: A conditional step now shows its conditions in the project language.
- Test sets: A sequential test set now runs in its execution order. A test set and a test plan keep their order after you move an item.
- Test runs: A run no longer switches to a tab that shows the browser error page of an HTTP error.
- Element detection: Thunders now keeps all variables of a click step. Thunders now finds elements in shadow DOM when their id has special characters.
- Thunders AI: Thunders AI tells you when the project blocks the browse location, so you know why it cannot browse. Long links in your messages now stay readable.
- Recorder: The activity list no longer shows duplicate rows for one recorder session.
- Squash TM: Thunders now sends the not run state to Squash TM as not run.
- UI polish: The environment screen now hides archived apps. The self-healing crop now shows the element in frames that are not 16:9. A conditional step that did not run now shows muted keywords. The dark mode links now have enough contrast.
Migration steps
- Pull the new images and redeploy, starting with the API Service (it applies any database updates automatically).
upgrade urgency
Recommended
Ready to ship faster with smarter testing?
