1. Highlight

    Featured this release

    PR review verdicts now weigh all of a PR's tests together

    The PR review agent assesses every test in a run side by side instead of one at a time, catching failures that share a root cause.

    Previously, each test in a PR review was assessed on its own. Now the same agent that reviews the PR looks at all of its test results together, so it can catch cases where several tests fail for the same underlying reason – a shared login step, for instance – instead of reporting them as unrelated failures.

    Tests still being assessed show as "Awaiting assessment" in the test composer while the agent works. This builds on the diff-anchored verdicts and self-repair work from the past week.

    # Direct link

  2. Update 02 of 06

    Search your entire Jira or Linear backlog from chat

    The chat assistant's tracker search is no longer capped at 25 issues – ask about your backlog and get up to 500 results back.

    If your project has Jira or Linear connected, asking the chat assistant to look up tracker issues now returns far more than a first page of 25. Large backlogs are searchable in one go, up to 500 issues at a time.

    # Direct link

  3. Update 03 of 06

    Clearer CORS failures in network logs

    Test run network logs surface preflight requests and the exact reason a CORS request was blocked.

    Preflight (OPTIONS) requests now appear in network logs with a "Preflight" badge and are searchable by that term. Requests blocked by CORS show a "CORS" label with a plain-language reason – like a missing Access-Control-Allow-Origin header – and the server's actual response headers, so you can see exactly why the browser rejected the call instead of a bare failed request.

    # Direct link

  4. Update 04 of 06

    Mobile build numbers on results, and downloadable preview builds

    Test results now show the exact build number, and preview mobile builds can be downloaded straight from the Preview Environments card.

    Mobile test results now display the build version and number (e.g. v1.2.3 · (42)) pulled straight from the uploaded APK/IPA, so you can pin a failure to the exact build.

    Preview environments created via CI/CD now show a platform badge, file name and size for mobile builds, with a download icon to grab the APK/IPA directly – matching the download experience already available on permanent environments. Promoting a preview environment to a custom one now carries its linked mobile build along with it.

    # Direct link

  5. Update 05 of 06

    Every test run now shows its outbound IP

    See the exact public IP address your target app saw during a run – handy for allowlist and firewall debugging.

    The run test case result page now has an "Outbound IP" field with a copy button, filled in live as the run executes. It pairs with the mobile testing IP ranges already available for allowlisting, but gives you the precise address for that specific run.

    Also exposed via the API as outboundIpAddress on run test case responses.

    # Direct link

  6. Update 06 of 06

    PR review chats now stay in one conversation

    Re-running a PR review continues in the existing conversation instead of starting a fresh chat every time.

    When @qatech reviews the same pull request again – after new commits, for example – it now reuses that PR's existing review conversation instead of spinning up a new chat each time, so the full PR review history stays in one thread.

    The change-review API behaves the same way; pass forceNewConversation: true if you want a clean conversation instead.

    # Direct link

Your team moves fast. Can your testing keep up?

QA.tech agents test your product autonomously, so moving fast never means shipping broken. See how it works in a 30-minute demo.

Get a demo