When a Test Run stops unexpectedly due to a system failure, it no longer hangs in an ambiguous state.

• Runs interrupted by a system failure are now marked with a clear Interrupted status.
• No more stale runs sitting in your list with no explanation.

You’ll always know what happened: whether a run passed, failed, or was interrupted before it could finish.

#Sprint26

A few targeted improvements to how Thunders Copilot works day-to-day.

• Interrupt in-flight requests: cancel a Copilot generation at any time if you change your mind.
• Variable highlighting: variables in Copilot responses are now visually highlighted, so you can spot dynamic values instantly.
• Autocomplete for assets and variables: your Test Assets and Project-level variables now appear in autocomplete when you type [.

These changes make working with Copilot faster and reduce mistakes when referencing variables or test assets.

#Sprint26

You can now export any test run as a full PDF report: steps, screenshots, results, everything in one document you can share.

• Custom introduction: add your own summary text at the top of the report.
• Branding: upload your company logo and customize report colors.
• Language: choose between English and French for the report language.
• Section control: toggle which sections appear from your project settings.
• Ready to share: export the final PDF for stakeholders, compliance, or audits.

This is currently in Preview. The exported PDF includes every step with its screenshot and result, so anyone reading it gets the full picture without needing a Thunders account.

#Sprint26

Flows that include native browser popups (alerts, confirms, prompts) used to break silently. Not anymore.

• Alert, confirm, and prompt dialogs are now fully supported during test execution.
• Thunders detects and handles JS dialogs triggered from any step, automatically.

If your app uses window.alert(), confirm(), or prompt(), Thunders will interact with those dialogs as part of the test flow. No workaround needed.

#Sprint26

In the last release, Thunders Copilot started thinking out loud. Now the AI Persona does too, directly inside your test results.

AI reasoning is now displayed in a collapsible section across the step list and bug reports. It covers:

• Successful steps and failed steps: see why the AI chose a specific action, what it detected on the page, and how it decided to proceed.
• Assertion steps: understand exactly what the AI checked, what it compared, and why it passed or failed.
• Conditional steps: follow the AI’s logic when it evaluates conditions and decides which branch to take.
• Expand or collapse per step, so reasoning is there when you need it and out of the way when you don’t.

When a step fails or does something unexpected, you no longer have to guess why. The AI tells you what it saw, what it considered, and what it decided, right in the recap.

#Sprint26

Test Sets used to be a way to group and run tests together. Now they work like real test campaigns, with execution control and live status tracking.

• Re-execute only failed tests from a campaign run, without re-running what already passed.
• Campaign status and pass rate update automatically after re-execution. No manual recalculation.
• Re-run individual test cases within a campaign, and the results attach to the same run with updated metrics.
• Works everywhere: from both the global Test Runs view and the Test Set detail view.

Previously, if a campaign had 3 failures out of 50 tests, you’d re-run the whole thing. Now you retry just those 3, and the campaign recalculates its pass rate on the spot.

#Sprint26

All list views across the platform now include Clear and Reset buttons in column filter popups, making it faster to manage and remove filters.

• Reset button: removes the active filter on a column and immediately refreshes the grid to show all rows again; available on all filter types (text, number, date, and set filters)
• Clear button: clears the filter input fields without closing the filter popup, letting you start a new filter from scratch; available on text, number, and date filters.

Sprint #25

RGAA 4 (Référentiel Général d'Amélioration de l'Accessibilité) is now available as an accessibility compliance target for your Accessibility Persona.

• New standard option: select RGAA 4 from the Accessibility Standard dropdown in the Accessibility Persona profile view
• French government compliance: RGAA 4 is the French national accessibility framework, required for public sector websites in France and widely adopted across French-speaking regions
• Joins the existing set of standards including WCAG 2.1 AA, WCAG 2.2 AA, WCAG 2.0 AAA, Section 508, and EN 301 549

Sprint #25

A new video-style player lets you review test run execution as a continuous, step-by-step visual replay instead of clicking through individual screenshots.

• Play/Pause/Replay controls: watch the entire test execution play back automatically, or pause at any point to inspect a specific moment
• Adjustable playback speed: choose from 0.5x, 1x, 2x, or 4x speed to review runs faster or slower
• Keyboard shortcuts: space to play/pause, arrow keys to navigate frames

Sprint #25

The Visual Compare step has been enhanced, giving you more control over how your UI is validated against reference designs.

• Full Page Audit mode: compare an uploaded reference image against the entire scrollable length of the live page, catching differences even below the fold
• Visible Viewport mode: compare the reference image against only what is currently visible in the browser window, useful for validating specific sections of a page
• Test results now clearly indicate Visual match or Visual mismatch with detailed reporting on any differences found

Sprint #25