Skip to main content

Katalon Studio Release Notes: Version 11.x

note
  • The in-app upgrade from 10.x to 11.x is not available. Visit Katalon website to download the latest 11.x version: Download Katalon Studio.
  • Core dependency updates:
    • Appium must be updated to v3.0+ for local mobile testing with Katalon Studio 11.
    • KRE v11+ requires OpenJDK 21. Ensure your environment is updated to Java 21 to avoid execution failures.

Version 11.0.1 (Latest Version)​

Release date: February 26, 2026

This release focused on stability and reliability following the 11.0.0 major release. This update introduces two new MCP Server tools: captureTestObject for generating Katalon-standard test object locators, and webUITakeScreenshot for capturing and displaying web page screenshots inline in StudioAssist. It also includes fixes across Smart Mobile Recorder, Web Recorder Plus, execution engine, Log Viewer, HTML reports, and KRE.

New features​

  • Added new MCP Server tools for StudioAssist:
    • captureTestObject tool to generate multiple locator strategies (CSS, XPath, ID, etc.) for a selected element and returns a Katalon-standard test object, using the same recording engine as Katalon Studio's built-in recorder.
    • WebUITakeScreenshot tool to capture a full-page or viewport screenshot of the current web page, save it to the project's Screenshots folder, and display it inline in the chat for both the user and the AI to reference during test authoring, debugging, or test object generation.

Enhancements​

  • Added support for rerunning tests with a new execution session, helping ensure clean environments between runs.
  • Restored deprecated MobileTestObject methods to maintain backward compatibility for existing mobile projects and scripts.
  • [Log Viewer UI] Improved UI for screen capture object and screen variable content to make captured information easier to review.
  • Adjusted how test listeners are listed in HTML reports to align with the new Log Viewer logic.

Fixes​

  • Katalon Studio crashed on Windows with an EXCEPTION_ACCESS_VIOLATION error.

  • Katalon Studio crashed when clicking a downloaded file in Chrome, and showed an error when using Chrome DevTools (Inspect).

  • [Intermittent] Keywords were not displayed in the Keywords Browser due to the error: NoClassDefFoundError: ZipFile.

  • Fixed browser‑based video recording so that videos are fully generated even when test scripts do not include an explicit Close browser step.

  • Corrected the getElementLeftPosition API so it respects the default locator (e.g., Attribute) instead of always using XPath.

  • Resolved problems where nested Shadow DOM elements did not work correctly in the new Execution View.

  • Fixed a UI issue where button hover states caused layout expansion unexpectedly.

  • Fixed an integration issue where reports could not be uploaded to qTest.

  • Fixed an issue where exported HTML reports were missing Profile information.

  • Test Case IDs were incorrectly displayed in the test case log section of HTML reports.

  • The email subject was blank when executing a test suite collection because the test suite name was not included.

  • A console log file was still generated even when the Console Log option was unchecked in Report settings.

  • Resolved issues that prevented browser‑based recording video from fully generating.

  • Copying a test case generated a duplicate .groovy file with the same filename, causing script mismatches when running in Debug mode.

  • Pasted objects could not be renamed due to XMLStreamException: ParseError + Invalid byte 1 of 1-byte UTF-8 sequence.

  • Windows test objects were missing hyperlinks in test step details, making navigation harder.

  • Protected variable values were removed after saving changes to a test case.

  • Execution tabs were not closed/cleaned up when closing a project.

  • Fixed issues preventing Web/Mobile AI Self‑Healing from working due to InjectionException: Could not find satisfiable constructor for ILlmService error.

  • Nested Shadow DOM elements could not be interacted with in the new Execution View.

  • NullPointerException warning when executing API Test Generation test cases with the SauceLabs plugin.

  • Test objects using XPath locators were loaded differently in KSE 11.0.0 compared to earlier versions, causing scripts that previously passed to fail when opened in the new version.

  • Mobile test objects that had XPath properties set at runtime threw the error XPath is null when using MobileTestObject instead of TestObject.

  • Test execution failed in KSE v11 with blank variable values even though data binding was correctly configured.

  • Execution with Custom Desired Capabilities failed with the error: NullPointerException: Name is null.

  • Execution in Remote mode with Selenium failed with the error: non-compliant desired capabilities (remoteMobileDriver).

  • [Smart Mobile Recorder]:

    • Recorder UI now behaves correctly on multiple monitors, and the device dialog properly shows action buttons.
    • Ensured dialog stacks correctly, so the Recorder dialog stays in front of the Device view and re‑run logs are visible.
    • Variables with null value can now be saved and no longer disappear from recorded scripts.
    • Fixed an error when saving variables with a null default value, preventing exceptions such as Cannot invoke "String.isEmpty()" because "this.defaultValue" is null.
    • Large values for Number variables no longer turn into Infinity or null when saved.
    • Custom keywords that used user-defined Enum types failed to load in the recorder.
    • Custom keywords added during recording are now correctly saved into test scripts instead of being removed.
    • Custom Desired Capabilities configured in Project Settings were not loaded when launching the Smart Mobile Recorder.

  • [MCP Server]:

    • Katalon Studio could not connect to the Atlassian MCP server.
    • Fixed a UI issue in the MCP Server connection dialog when connecting to servers that require authentication.

  • Fix multiple issues with Katalon Runtime Engine:

    • KRE 11 did not recognize custom or remote browser types, causing execution to fail.
    • Unable to continue test execution using an existing browser session.
    • The -testOps.baselineCollectionId argument was not recognized when executing tests with KRE.
    • [macOS] KRE failed to run when executed from the same directory as the binary using the ./katalonc command.

  • [Web Recorder Plus]:

    • Could not save/playback when using Open Browser with an undefined URL.
    • Running logs were lost when switching to other tabs (Captured Objects / Variables).
    • Recording with an active browser paused unexpectedly.
    • The browser session was no longer available after a playback run, preventing further recording.
    • Unabled to record with Edge Chromium.
    • The browser type logo was hidden in Active Browser mode when the page title was too long.
    • Keyword utility (KeywordUtil) methods failed to load when continuing a recording on an existing test case.
    • The ENTER key was not recognized when playing back recorded steps.
    • Continuing a recording on an existing script required manually starting a new browser session instead of resuming automatically.

  • [Web Recording Agent]:

    • Fixed an issue where generated steps were not cleaned up after users selected “Do not show again”.
    • The Add, Run, and Settings buttons remained active while test step generation was in progress, allowing unintended interactions.
    • The Web Recorder showed an empty state when appending an existing test case that contained custom keywords.
    • The background task plan continued running even after the user manually terminated the step generation process.

  • [Log Viewer, Job Progress UI]:

    • Fixed an intermittent bug where the Log Viewer showed a persistent “Loading” icon for data binding tests and test suites.
    • Fixed Job Progress UI for terminated tests so their status is visible again.
    • Updated Job Progress so test suite names are no longer displayed in red incorrectly.
    • Ensured that previous execution logs are cleared when closing a project.
    • Fixed issues where the Log Viewer did not auto-scroll to the current running step.
    • Fixed hyperlinks issue for Windows test objects in test step details.

  • [StudioAssist]:

    • Fixed the test object icon in the Modified Files list so it is consistent with the icon in the Project tree.
    • Resolved an issue where the current file disappeared after clearing the conversation and attaching a new file.
    • Corrected the current file selection state so that the UI always reflects the actual selected file, including when it was not selected in the previous session.

Version 11.0.0​

Release date: January 26, 2026

Overview​

This release delivers a major platform and dependency upgrade, introduces official support for Recorder Plus and Automated Code Migration tool, adds powerful new capabilities such as Smart Mobile Recorder, Katalon Studio Tracer, StudioAssist Agent profile, and new MCP tools. It also includes UI enhancements and extensive fixes across StudioAssist, test execution, WebUI and mobile testing, significantly improving overall performance, stability, and reliability.

New features​

  • Upgraded from Java 17 to Java Eclipse Temurin 21 LTS. Linux users must ensure OpenJDK 21 is installed in their execution environment.
  • Upgraded to Selenium 4.39.
  • Upgraded Appium Java Client to 10.0.0.
  • Upgraded Cucumber runtime to version 7.
  • Introduced AI Mobile Self-healing - an AI solution to suggest alternative locators when existing mobile test object locators fail, using current locator values and metadata. See AI self-healing.
  • Introduced Smart Mobile Recorder - an upgraded experience for recording mobile app interactions. The feature allows you to capture user live actions, and generate test steps automatically. Learn more at Smart Mobile Recorder.
  • Introduced Katalon Studio Tracer - a trace-recording capability that captures DOM snapshots, network activity, console logs, and browser state to help debug test failures and flaky behavior. Learn more at Katalon Studio Tracer.
  • Introduced AI Recording Agent (beta) in Web Recorder Plus - converts natural-language scenarios into structured test cases and allows seamless switching between AI-driven and manual recording within a single session. Learn more at Generate a Web UI test case with AI.
  • [StudioAssist | Agent mode]:
    • Added a new MCP toolset for WebUI testing. See WebUI tools.
    • Added updateInternalTestData and getInternalTestDataDetails to Test Data MCP toolset.
    • Introduced Agent profile - where you can add custom instructions for StudioAssist to apply in a specific testing scenario. This feature allows team that shares a Katalon Studio project to get consistent AI responses aligned with their test framework. Learn more:
  • Added new Mobile keywords:
  • API test generation and report viewing with AI - Import your OpenAPI file to have Katalon’s AI agent automatically generates tests, executes them, and provides a results report. Learn more at API test generation and report viewing with AI.

Enhancements​

  • Introduced a new and enhanced look for Log Viewer and Job Progess: Cleaner, more accurate logs that better reflect real execution behavior and help users navigate failure points faster. (attached images)
    • Added an Execution Viewer toggle in Project Settings to enable or disable the new UI.
    • Refined Log Viewer to make debugging and analysis clearer.
    • Removed generic or misleading error messages when no root cause is found.
    • Updated test execution architecture for improved performance.
  • Improved variable binding by enabling one-click application of test-case-level binding across all test cases in a test suite.
  • [StudioAssist] Added the ability to review, undo file changes, and view them in a diff view. Learn more at Review and revert file changes.
  • Updated onboarding flow: removed initial welcome dialog and disabled Web Recording Tour when using Katalon Web Recorder Plus.
  • Added Chrome 144, Edge 144, and Firefox 147 compatibilities.
  • Added new migration recipe Migrate from Studio 10.x to 11.x to help automatically migrate projects from version 10 to 11, including Cucumber 3 to Cucumber 7 migration.
  • [Security Compliance]:
    • Upgraded poi-ooxml-3.17.jar to version 5.2.3.
    • Removed unused BouncyCastle cryptography library bundle.

Changes​

  • Web Recorder Plus is now officially supported for general users.
  • Auto-migrate scripts now officially supported.
  • [UI] Moved AI Configuration to a separate top-level section in Katalon Studio Settings instead of being located within StudioAssist.
  • [UI] Re-organized components inside StudioAssist chat window.
  • [UI] Web Recorder Plus settings is moved from Beta Features section to Recorder.
  • Removed the redundant option of BDD: Run as 1 Cucumber Feature.
  • No longer provides a separate free package of Katalon Studio. Free users can continue use the same package after trial ends.
  • When creating new test suites, Test Case-level variable binding is now enabled by default, so your test suites work immediately without additional configuration. To use suite-level binding instead, uncheck "Use Variables and Binding at Test Case" and check "Use Variables and Binding at Test Suite".
  • [Web Service] Removed support for OAuth 1.0 Authentication. If your web services use OAuth 1.0, migrate to OAuth 2.0 authentication.
  • [Web Service] Removed support for WADL import.
  • Replaced the SWT browser with Equo Chromium for internal web views, ensuring more consistent rendering and behavior across Windows, macOS, and Linux.
  • As gpt-4.1-mini will soon be deprecated, Katalon AI Service is upgraded to use gpt-5-mini model.
  • Execution profile fallback logic: If no profile is specified for execution, the system now falls back to the profile explicitly marked as "Set as default execution profile".

Fixes​

  • [Mobile] Unable to perform the Click action on Android devices.
  • [Mobile] Fixed an issue where tap actions were executed on the wrong element.
  • [Mobile] Unable to use keyword in the Browserstack remote platform.
  • [Mobile] The Tap and Hold input dialog still executed the action even when users clicked Cancel.
  • [Mobile] The Test Object locator was not updated after edits were made to the object properties.
  • Error while creating project and importing API.
  • The validateOpenAPIAgainstSpecification keyword did not work as expected.
  • Large BDD projects experienced poor performance during the Loading BDD scripts step.
  • Error thrown when closing or switching between BDD projects.
  • [StudioAssist]:
    • Chatbot occasionally became blank after users submitted a question, causing the question to disappear.
    • When Auto-tagging was enabled and managed by Admin, the Generate Code feature would fail with an error message "AiAutoTagSettingProvider is not set".
  • Smart Wait did not work when running via Run or Debug from here.
  • When running test suites, global variables would not load correctly if a non-default execution profile had been set as the default profile.
  • Added automatic rerun file generation for failed BDD scenarios, enabling selective rerun of failed Cucumber tests.
  • You can now specify packages for step definitions in Project Settings > Cucumber instead of using CucumberKW.GLUE. Also added a feature file scanning option to choose between Scan on project open or Scan on file open.
  • Websites failed to load or only partially loaded when Katalon Smart Wait Utility or Katalon Studio extensions were enabled.
  • Retry mechanisms did not work when executing a test suite collection in KRE.
  • Keyword proposal in Script view did not work.
  • [AI Self-healing]:
    • Could not find satisfiable constructor in com.kms.katalon.ai.core.services.ILlmService error when using AI self-healing.
    • KSE showed a long path error even though the previous run completed without issues.
  • [UI] Multiple help icons were displayed on Web Service requests, dynamic test suites, and test suite collection editors.
  • [UI][macOS] The "Click here to fix the broken test object" link did not appear in the Log Viewer when a test failed with a WebElementNotFoundException.
  • [macOS Tahoe 26.1] When selecting multiple lines of text in the Script Editor, only the first line appeared as selected.
  • [macOS] An "internal error occurred during: Indexing project" error appeared when opening projects.
  • Users were unable to login to Katalon Studio with SSO because the login dialog would remain open even after successful authentication to TestOps, blocking access to the application.
  • [Spy Plus] Switching between Chrome and Edge browsers would cause Spy Plus to stop capturing test objects.
  • [Spy Plus] Entering a different website URL during recording or spying would cause Spy Plus to insert the "close browser" command at an incorrect position.
  • [Recorder Plus] When using Active Browsers mode, recording sessions would continue capturing actions even after clicking the stop button.
  • [Web Recorder] The navigateToURL keyword was not automatically added after openBrowser when recording on a new browser for certain web pages.
  • The editor froze when opening large Groovy files (10,000+ lines) while in debug mode.
  • [Windows] The editor frequently hung when switching to script mode and typing "do" without quotation marks.
  • When copying and pasting a step with an input value in the Web Recorder and then changing that input value, the object path in the saved script was missing the folder path.

Known limitations​

  • Due to incompatibility between Sikuli 1.0.2 and Java 21, using WebUI.click and WebUI.setText on an image will return an error.
  • [BDD testing]:
    • Background steps no longer appear in Cucumber HTML reports due to Cucumber 7's implementation.
    • Test-run-level hooks information does not appear in Cucumber reports. The Hooks section in reports remains empty even when test-run-level hooks are executed.
  • [WebUI MCP tools]:
    • Cannot run multiple concurrent browser sessions on Safari.
    • The webUITakeAccessibilityTreeSnapshot tool is not available when working with Safari.
  • [AI self-healing]:
    • AI self-healing might have issues finding the element with the image locator.
    • In some cases, AI Self-Healing may not be triggered promptly when standard self-healing fails.
  • [UI] Button hover causes layout expansion and visual misalignment.
  • [Spy] Selected locator behaves incorrectly when finding element by attributes.
  • [Web Recorder Plus]:
    • Changes are not saved when disabling Enable Smart Locator option in Web Recorder Engine Settings.
    • Unable to record with Edge Chromium. Workaround: Go to Katalon Studio > Settings... > Recorder and click Apply and Close again, then retry.
    • Some scenarios may still record or replay the ENTER key incorrectly, either as plain text or not at all on certain pages.
    • Test cases that use advanced keywords (e.g., KeywordUtil or custom keywords) or interact with complex DOM structures might behave differently between Recorder Plus playback and normal Run.
    • NULL-type variables with null default values are silently dropped when saving the script.
    • When saving, some variables have their type unexpectedly changed to Method call or trigger an “invalid test case” error.
    • Number-type variables with very large values are converted to “infinity” and then show as null on save.
    • In some situations, Active Browser mode may not behave as expected:
      • Playback can open a different browser than the one being recorded.
      • Recording may start in a paused state or not expose the new browser session for reuse.
      • Browser icon may be hidden for long page titles in the Active Browser list.
    • In some environments, Recorder Plus can experience intermittent connectivity or compatibility issues.
    • Continuing recording on an existing test case or script may not behave consistently.
    • Some recorded scripts may still fail to resolve GlobalVariable values correctly when executed under non-default profiles during Recorder Plus playback.
    • Some usability issues may still affect the Recorder Plus UI during recording and playback:
      • Newly recorded steps may not run correctly in certain flows.
      • Recorder Settings panel may not auto-collapse when starting an Active Browser session.
      • Running Log may disappear when switching between tabs during playback.
  • [Web Recording Agent]:
    • When using Web AI Recording Agent, the agent may fail to generate test steps and instead display only an error message.
    • Stopping or terminating the test step generation process does not cancel the underlying task plan as expected.
    • After expanding the Recorder Settings panel, the AI prompt field cannot be edited.
  • In Web Recorder Plus and Smart Mobile Recorder, custom keywords added during a recording session were not saved to the generated test case and could be removed automatically.
  • When an error occurs in the task plan, subsequent steps are still displayed instead of being hidden.
  • Web RA’s interface shows inconsistent font sizes and a misaligned or inconsistent settings icon.
  • Web RA may generate steps that use incorrect or mismatched keywords.
  • When choosing “Do not show again” for certain Web RA prompts, previously generated steps are not cleaned up as expected.
  • While Web RA is generating test steps, the Add / Run / Settings buttons remain enabled instead of being temporarily disabled.
  • [Smart Mobile Recorder]:
    • When saving, some variables have their type unexpectedly changed to Method call or trigger an “invalid test case” error.
    • Number-type variables with very large values are converted to “infinity” and then show as null on save.
    • If the default locator doesn’t match any element, using Image locator’s Take Screenshot throws an error.
    • Saving a test case that includes a NULL-type variable (with default value NULL) throws a Java error about defaultValue being null and stops saving.
    • NULL-type variables with null default values are silently dropped when saving the script.
    • On iOS (UIKitCatalog), recorded tap steps on list items (XCUIElementTypeStaticText) show as recorded but no actual tap occurs on the simulator.
    • In iOS (UIKitCatalog), Interactive mode taps sometimes hit the wrong element (e.g., tapping Slider section actually triggers ProgressView). This can also cause capturing the same object multiple times.
    • In iOS (UIKitCatalog 2.13), the highlight rectangle is drawn in the wrong position when hovering/moving the cursor over test objects.
    • Old Mobile Recorder shows outdated iOS device versions when using TestCloud devices.
    • Using Verify and Highlight with a “Custom locator” throws an error indicating that custom locator isn’t supported.
    • When starting with TestCloud, if the Appium-device connection is slow or times out, the Recorder dialog appears but the Device View never shows. No clear timeout error is presented.
    • Recorder UI does not reflect in the 2nd monitor. Device dialog cannot display the action button.
    • In Interactive mode, when an in-app dialog is shown, tapping outside it (which should dismiss the dialog) times out and no step is recorded—even though in emulator that tap would close the dialog.
    • If a locator matches multiple elements, neither Interactive mode nor manual mode performs a click.
    • In Interactive mode, function keys such as Shift, Tab, etc. get sent into the text field as actual input.
    • [UI] When using Image locator → Take Screenshot → Highlight, an error appears (since Image locator isn’t supported yet), but the error message text overflows/overlaps the dialog bounds.
    • [UI] The Device dialog appears too large with lots of unused white space by default.
    • getElement keywords (getElementLeftPosition, getElementTopPosition, getElementWidth, getElementHeight) always attempted to locate elements using XPath, even when the Test Object was configured to use Attribute as the default locator.
    • For “Tap and Hold” duration input, the dialog allows choosing complex types (Global Variable, Variable, etc.), but the recorder doesn’t support them, resulting in recorded steps always having zero duration.
  • [Katalon Studio Tracer]:
    • Help text for katalon -help trace is missing, so users cannot see the documented list of Tracer CLI parameters and their usage when configuring Tracer from the command line.
    • When using on first retry mode with more than one retry, the Tracer icon only appears on the retry entries, not on the main run in the Katalon Studio report view, making it harder for users to discover that a trace is available for that test.
    • DOM snapshots in the Tracer file fail to load correctly, so users cannot view page state snapshots for each step even when this data capture option is enabled.
    • For failed test cases that throw errors, the Error tab in the Tracer view remains empty, preventing users from seeing captured error details as part of the trace.
    • Browser information is not captured in the Tracer metadata, so users cannot easily identify which browser environment a trace was recorded in.
    • When using callTestCase, the Tracer file only captures the before steps and omits the main actions from the called test, resulting in incomplete execution traces.
    • For incomplete test cases (for example, those calling System.exit(0)), no Tracer file is created, so aborted or incomplete executions are not traceable.
    • In on failure or on first failure modes, blank Tracer files are still created for skipped tests, generating unnecessary trace artifacts with no useful data.
  • [Log Viewer]:
    • History view, Jump to issue and Troubleshoot are not yet supported in the new execution viewer.
    • In some executions, the Log Viewer may not auto-scroll to the currently running step (especially with long logs or BDD tests), you might need to manually track progress.
    • For tests without a Start browser/Start app step, the Execution Information section may not be shown in Log Viewer.
    • After closing a project, previous execution logs in Job Progress / Log Viewer / Console might remain visible until a new execution is triggered.
    • [Intermittent] For some data-binding test suites, the Log Viewer can stay in a “Loading” state for certain test cases even after the execution finishes.
    • Steps using markWarning() might be displayed with warning messages treated as failure reasons.
    • For some executions running in headless modes or custom profiles, the Run With mode and Profile used may not appear in Log Viewer.
    • In conditional flows, the else branch might not be correctly represented/count in the Log Viewer tree.
    • The embedded Log Viewer may occasionally log JavaScript errors, which can potentially impact how tree data is rendered in the UI.
    • In tree view, only info logs may be displayed under keyword child nodes, while warning and error logs might not appear in this view even though they exist in the execution.
    • In Windows test executions, test object references in step details may appear as plain text instead of hyperlinks.
  • Unable to upload test results from Katalon Studio to qTest.
Was this page helpful?