Skip to main content

Katalon Studio Release Notes: Version 11.x

note
  • In-app upgrade is not available when upgrading from the following versions:
    • Katalon Studio 10.x to 11.x
    • Katalon Studio 11.0.0
    • Katalon Studio 11.0.1
      To upgrade from an affected version, download the app from Katalon website.
  • Core dependency updates:
    • Appium must be updated to v3.0+ for local mobile testing with Katalon Studio 11.
    • Katalon Test Execution - Local v11+ requires OpenJDK 21. Ensure your environment is updated to Java 21 to avoid execution failures.

Version 11.2.1 (Latest Version)​

Release date: June 15, 2026

This release introduces new users to guided onboarding for AI-assisted test creation and web test recording with Recorder Plus. It also includes usability improvements, compatibility updates, and fixes that make Studio more reliable across common testing workflows.

New features​

Create tests with AI onboarding tour​

  • What's new: First-time trial users can use the new Create test with AI onboarding tour to create and run their first web test. Studio helps create a starter project, collect the application URL, and open Katalon AI Assistant in Agent mode.
  • Why this matters: New users can see a working automated test sooner, without learning the full Studio interface or writing code first.
  • How it works: Follow the Create test with AI onboarding tour when starting Katalon Studio as a first-time trial user. Alternatively, navigate to Knowledge Hub and select the tour.

Changes​

  • Upgraded Eclipse platform from 2025-12 to 2026-03.

Enhancements​

Enhancements
  • When Web Recorder Plus is enabled in Katalon Studio Preferences > Recorder, the Web test recording onboarding tour uses Recorder Plus. This gives users a guided recording experience that matches their selected recorder configuration.
  • Studio now provides clearer links from Katalon Studio to Katalon True Platform. Entry points have been cleaned up, and project links now point users to relevant True Platform or TestOps Legacy destination.
  • You can now easily access True Platform execution result directly from Job Progress or via the get execution result MCP tool after a run finishes.
  • Studio now checks the online status of API endpoints that use the GET method, improving compatibility with more endpoint configurations.
  • Katalon AI Assistant settings now explain where each project context option applies. Object Repository and Custom Keywords apply to Ask mode, Generate Code, and Explain Code. Follow-up question suggestions apply to Ask mode.
  • Upgraded default model for Katalon AI Service to gpt-5.4.
  • Improved Start Page UI with a cleaner layout, refreshed True Platform banner, and a dedicated News panel for release highlights.
  • Added Chrome and Microsoft Edge version 149 browser compatibilities.
  • [Security Compliances]:
    • Upgraded BouncyCastle to 1.83.
    • Upgraded PostgreSQL to 42.7.11.
    • Upgraded Apache Log4j 1.x Compatibility API to 2.25.4.

Fixes​

Fixes
  • The Knowledge Hub and onboarding screens now display correctly on Ubuntu. Previously, parts of the onboarding UI rendered incorrectly or block the test suite onboarding tour.
  • Offline license expiration dates now match between License Management and Katalon Studio. Previously, Studio could show the wrong expiration date because of a timestamp format mismatch.
  • Azure DevOps uploads now support special characters in test case error messages. Previously, uploads could fail with an HTTP 500 error when Azure DevOps received malformed character encoding.
  • REST request body type selection now saves correctly when users switch between JSON and XML. Previously, saving the request could revert the body type to the previous option.
  • KeywordUtil.markFailed() now marks BDD steps and scenarios as failed correctly after the Cucumber 7 upgrade. Previously, affected BDD executions could continue without failing the step or scenario.
  • Katalon AI Assistant file attachments now upload correctly behind corporate proxies. Previously, blocked uploads could return an "Object not found" error.
  • Smart Mobile Recorder now handles scroll-to-text actions more reliably after scrolling down to target text. Previously, the recorder might not scroll back up as expected.
  • Test Execution Cloud runs from Studio now load available cloud devices and applications correctly. Previously, the Test Execution Cloud dialog could open with empty dropdowns.
  • Firefox WebDriver updates now complete successfully from Tools > Update WebDrivers > Firefox. Previously, GeckoDriver updates could fail with a candidate URL error.

Version 11.2.0​

Release date: May 28, 2026

Deprecations​

In this release, the following Katalon Studio MCP tools are deprecated and replaced.

Deprecated toolReplaced by
createTestCase, updateTestCasewriteTestCase
getTestCaseScriptreadFile
updateTestCaseScriptwriteTestCase
getTestCaseslistEntityFiles
createTestSuite, updateTestSuitewriteTestSuite
createDynamicTestSuite, updateDynamicTestSuitewriteDynamicTestSuite
createTestSuiteCollection, updateTestSuiteCollectionwriteTestSuiteCollection
getTestSuitereadFile
createWebTestObject, updateWebTestObjectwriteWebTestObject
createMobileTestObject, updateMobileTestObjectwriteMobileTestObject
createWindowsTestObject, updateWindowsTestObjectwriteWindowsTestObject
createRestWebServiceTestObject, updateRestWebServiceTestObjectwriteRestWebServiceTestObject
createSoapWebServiceTestObject, updateSoapWebServiceTestObjectwriteSoapWebServiceTestObject
getTestObjects, getTestObjectlistEntityFiles and readFile
createCustomKeyword, updateCustomKeywordwriteCustomKeyword
getCustomKeywordreadFile
getCustomKeywordsgetCustomKeywordMethods. This replacement returns method signatures instead of a class list.
createBddFeatureFile, updateBddFeatureFilewriteBddFeatureFile
createInternalTestData, updateInternalTestDatawriteInternalTestData
getTestDataListlistEntityFiles
getInternalTestDataDetailsreadFile

New features​

Multiple chat conversations and history in Katalon AI Assistant​

  • What's new: Katalon AI Assistant now supports multiple separate chat conversations. A history panel lets you revisit and switch between past sessions at any time.
  • Why this matters: Previously, starting a new question meant losing your previous conversation context. With multiple conversations and history, you can maintain parallel work streams and return to earlier chats without losing context.
  • How it works: Click the All chat menu on the upper left corner to open the chat history. Refer to the short walkthrough video below.

Smart Mobile Recorder — Advanced Configuration screen​

  • What's new: Smart Mobile Recorder now includes an Advanced Configuration screen where you can review, customize, and fine-tune how captured objects are identified, including managing attributes and locator strategies before saving.
  • Why this matters: Previously, Smart Mobile Recorder applied fixed object identification settings with no way to adjust them during recording. The Advanced Configuration screen allows you to make live edits to the test object's locator, reducing manual correction work afterward.
  • How it works: While recording, switch to the Captured Objects tab and select a test object to trigger the Advanced Configuration option where you can review and adjust object attributes and locator strategies before saving. See the walkthrough below:

Web Recorder Plus — Advanced Configuration screen​

  • What's new: The new web recorder (R+) now includes an Advanced Configuration screen for captured objects. You can review, edit, and adjust object attributes and locator strategies directly within the recording session before saving.
  • Why this matters: The Advanced Configuration screen allows you to make live edits to the test object's locator, reducing manual correction work afterward.
  • How it works: While recording, switch to the Captured Objects tab and select a test object to trigger the Advanced Configuration option where you can review and adjust object attributes and locator strategies before saving. See the walkthrough below:

Enhancements​

Enhancements

  • Drag and drop attachments to Katalon AI Assistant. You can now drag and drop files directly into the AI Assistant chat window to attach them, in addition to using the attachment button.

  • Failure Category Classification in AI Failure Analysis reports. HTML and email reports generated by AI Failure Analysis now include a Failure Category Classification section, grouping failures by category so you can identify patterns across a test run at a glance.

  • TestCloud Tunnel support for Mobile Native App. Previously, TestCloud tunnel supports private/local testing only for desktop browsers and mobile browsers. With the added support for mobile native app testing, this closes the last gap in site testing coverage.

  • AI-generated indicator on Katalon AI Assistant artifacts. Artifacts such as test cases, test suites and keywords generated by Katalon AI Assistant now display an AI-generated icon, making it easy to distinguish AI-created content from manually written content.

  • Edit keywords directly in the Recorder. You can now edit keywords — including renaming and modifying parameters — directly within the recording session in both R+ and Smart Mobile Recorder, without leaving the recording flow.

  • Resizable panels in Web Recorder Plus. The Running log and test step panels can now be resized by dragging the divider between them, letting you adjust the layout to match your workflow.

  • Recorder dialog auto-expands for existing test cases. When you open the Recorder with a test case that already has recorded steps, the recorder dialog now automatically expands to show those existing steps, giving you full context immediately.

  • Variable binding state indicators in Call Test Case. The Call Test Case input dialog and test step manual view now show binding state indicators for each variable — matched, unmatched, missing, or inherited — making it immediately clear which variables are correctly linked to the called test case and which need attention.

  • Edit Value Type for parameters directly in Recorder steps. A pencil icon now appears on parameter fields in R+ and Smart Mobile Recorder recorded steps, letting you change the Value Type for a parameter without leaving the Recorded Steps panel.

  • KS MCP tools consolidated. The Katalon Studio MCP tool set has been consolidated into fewer, better-described tools, improving usability when integrating Katalon Studio with AI assistants via MCP. See the Deprecations section for the full list of replaced tools.

  • AI Self-Healing now prioritizes element screenshot for locator identification. When AI Self-Healing triggers, it now prioritizes the element screenshot when searching for the correct locator. This improves healing accuracy for elements whose text or attributes have changed but whose visual appearance remains stable.

  • Added Chrome 148 and Edge 148 browser compatibilities.

  • Katalon AI Assistant chat history auto-condensation. The AI Assistant now automatically condenses older chat history to keep long sessions within context limits, so extended conversations remain responsive without requiring you to manually clear the chat.

  • Web Recording Agent UX improvements. Several usability issues in the Web Recording Agent have been resolved: the Settings panel now has correct focus and is editable when opened, and additional interface inconsistencies have been addressed.

  • Recording Agent edit mode: Cancel behavior corrected and primary CTA renamed. In Recording Agent edit mode, pressing Cancel now correctly discards unsaved changes and returns to the previous state. The primary call-to-action button has been renamed for clarity.

  • Web recording context menu now uses updated assert keywords. The context menu in the web recording engine now maps recorded assertions to the current assert keyword set, ensuring recorded assertions use the correct, up-to-date keywords on playback.

  • Temporary Groovy scripts in Libs folder now cleaned up automatically. Temporary Groovy scripts that previously accumulated in the Libs folder across multiple executions are now removed each time a new execution starts, keeping your project folder clean.

  • Updated AI Configuration prefilled defaults. The prefilled default values in the AI Configuration panel have been updated for this release.

Fixes​

Fixes

  • KRE on macOS: katalonc now has the correct execute permission. The katalonc executable in Katalon Runtime Engine for macOS now ships with the correct file permission and can be launched directly. Previously, running KRE on macOS failed immediately with a "no permission" error.

  • KRE 11.1.3 regression resolved: "Error while parsing selector" no longer thrown. Katalon Runtime Engine no longer throws a JavaScript "Error while parsing selector" error during test execution. This error was introduced as a regression in KRE 11.1.3 and was not present in KRE 10.4.3 or earlier.

  • Map-type profile values can now be saved. Map-type variable values in execution profiles can now be saved correctly in Katalon Studio 11.2.0. Previously, saving a profile that contained a map value silently failed or threw an error.

  • Browser-based recording now works with Chrome and Edge Chromium 148. Recording sessions using Chrome and Edge Chromium 148 no longer fail. Previously, a breaking change in the Chrome DevTools Protocol in version 148 caused a "CDP no-op DevToolsException" that prevented recording from starting.

  • Web Recording Agent: "Replace all current steps" checkbox now appears correctly. When continuing a recording session on an existing test case, the "Replace all current steps" checkbox now appears as expected. Previously, the checkbox was missing when continuing recording on an existing test case.

  • Login with proxy now works correctly — regression from v11.1.2 resolved. Logging in to Katalon Studio when a proxy is configured now works correctly. This was a regression introduced in v11.1.2; v11.0.1 and earlier were not affected.

  • Katalon AI Assistant: AWS Bedrock certificate error resolved. The AI Assistant no longer throws an "AWS Bedrock: Unable to execute HTTP request: certificate_unknown" error. Previously, this error prevented StudioAssist from functioning in certain network environments.

  • Console log: error messages are now easier to read. Error entries in the Console log tab are now displayed in a more readable format.

  • Masked variable values no longer exposed in execution log. Masked test case variable values are no longer visible in the execution log when one test case calls another. Previously, the actual value of a masked variable appeared in the log output of the calling test case.

  • Renaming a Keyword class now works correctly — regression from KS 10.4.2 resolved. Renaming a Keyword class in Katalon Studio now correctly renames all associated files and references. Previously, the rename appeared to succeed but silently left the class name unchanged — a regression introduced in KS 11.1.1.

  • Katalon Studio no longer crashes when opening the Java documentation viewer. Viewing Java Doc in Katalon Studio 11 no longer causes a crash. Previously, opening the Java Doc viewer triggered a crash related to the Equo CEF component in KSE.

  • [Linux] Editor View regression resolved. On Linux, the Editor View for Test Cases, Test Objects, and other editors is now correctly displayed and interactive. Previously, a regression caused the entire editor area to be blank or unresponsive on Linux installations.

  • Smart Mobile Recorder: objects can now be captured while the iOS keyboard is visible. Smart Mobile Recorder can now capture objects correctly when the iOS system keyboard is displayed on the device. Previously, the presence of the iOS keyboard blocked all object capture during recording.

  • Web Recording Agent: Scroll and SendKeys steps now use the correct keywords. Scroll and SendKeys actions recorded by the Web Recording Agent now map to the correct keywords on playback. Previously, these recorded steps referenced mismatched keywords that produced incorrect behavior during playback.

  • Test Object editor: switching locator no longer auto-selects all attributes. Changing the locator strategy in the Test Object editor no longer selects all attributes automatically. Previously, a regression caused every attribute to be selected whenever the locator type was switched.

  • Ubuntu: layout no longer jitters or flickers. The Katalon Studio interface on Ubuntu no longer displays jitter or flickering during normal use.

  • waitForElementNotVisible now works as expected. The waitForElementNotVisible keyword now correctly waits for the target element to become invisible before returning.

  • Fixed an issue where TestCloud test cases failed with error SessionNotCreatedException: Account ID in access token does not match the request thrown.

  • KeywordUtil.markFailedAndStop() now correctly marks both the test case and test step as failed. When KeywordUtil.markFailedAndStop() is called, the test step status now matches the test case status and shows as failed. Previously, the test case was marked failed but the individual test step displayed as passed.

  • Test Run panel now shows the correct execution ID and valid TestOps link. The Test Run panel now displays the correct execution ID, and the deep link navigates to the correct execution page on TestOps. Previously, the panel showed a wrong execution ID and the deep link redirected to an invalid page.

  • KRE no longer logs "Illegal character in path" warning. Katalon Runtime Engine no longer generates a warning about "Failed to get katalon.ini location: Illegal character in path at index 78" during startup.

  • Masked variables are now hidden during playback in R+ and Smart Mobile Recorder. Masked variable values are correctly hidden in the running log during playback in the new web recorder and Smart Mobile Recorder. Previously, masked variables displayed their actual value in the log during playback.

  • Clicking the "How to set up Katalon Test Execution Cloud Tunnel?" link in the configuration dialog no longer produces an error.

Version 11.1.3​

Release date: May 06, 2026

This release addresses Smart Locator execution regressions, proxy login failures, Web Test Object display issues and Katalon Studio Copilot upgrade routing.

Fixes​

  • WaitForElementNotPresent() now works with Smart Locator during test execution. Previously, tests could fail with a Groovy runtime error when this keyword used a Smart Locator.
  • Login now works when Katalon Studio is configured to use a proxy. Previously in v11.1.2, users could not log in through a configured proxy.
  • Web Test Objects now display and behave correctly. Previously, some Web Test Objects could be mishandled as Mobile objects.
  • In-app upgrades now download the correct Copilot package for Katalon Studio Copilot. Previously, Copilot builds could upgrade to the non-Copilot package.

Version 11.1.2​

Release date: April 23, 2026

This release focuses on expanding agentic execution in Katalon AI Assistant, improving reliability across recording and execution workflows, delivering usability updates for reporting, Git project handling, self-healing, and addressing a broad set of user-reported fixes.

Deprecations​

  • -reportFolder is deprecated and replaced with -report.folderPath for correct Test Suite Collection report folder structure.
    • If you have existing execution commands using -reportFolder, update them to -report.folderPath before running them in version 11.1.2.

Changes​

  • Agent mode is now the the default mode when launching Katalon AI Assistant.
  • [AI self-healing] The Page Source option in Input data sources settings is now enabled and cannot be unchecked when AI self-healing is enabled. This ensures the AI model always has the context it needs to generate valid locator suggestions.
  • Changed UI text from "Show Start Page contents" to "Show start page at startup" in Katalon Studio Preferences.

New features​

MCP Execution tool in Katalon AI Assistant​

  • What's new: Added the execution tool set that allows Katalon AI Assistant to execute a test case, test suite, or test suite collection in Agent mode.
    • executeTest
    • getAvailableMobileDevices
    • getAvailableTestCloudEnvironments
    • getTestExecutionResults
    • getTestExecutionStatus
    • stopTestExecution
  • Why this matters: With the execution tool set, the full agentic loop is now possible. You create, run, fix, and re-run tests without leaving the agent.

For more details, refer to Execution tools.

[Web RA] Continuous Prompting with Context Awareness​

  • What's new:
    • You can now generate new steps on top of generated steps, in the current recording session or an existing test case.
    • Added a checkbox to Replace all current generated test steps, if selected the generated steps will overwrite existing steps.
  • Why this matters: Previously, users had to re-prompt from scratch or manually clean up steps when refining generated flows. Continuous prompting keeps context across iterations, so you can extend or replace steps faster with less rework and fewer prompt retries.
web recording agent generate over existing steps

takeFullPageScreenshotWithScroll() WebUI keyword​

Enhancements​

Enhancements
  • Improved the existing Git clone workflow to handle multi-project repositories properly. Each .prj is treated as a separate project, and the first one found opens automatically. Default folders (Test Cases, Test Suites, etc.) are no longer created at the repository root.
  • [Job Progress] You can now review completed and in-progress executions in the Job Progress window. Job progress window with history view
  • Added a [AI Self-Healing] prefix for entries that performed the healing in Log Viewer to improve traceability and troubleshooting.
  • Start time in test execution email reports now include the timezone to reduce confusion when executions ran in different time zones, applying to Test Suite and Test Suite Collection emails.
  • Improved Katalon AI Assistant error messaging when users exceed the MCP tool limit of 128 tools.
  • Enhanced Katalon AI Assistant to correctly track attachment context even when you switch between attachments during a conversation.
  • [MCP] Improved WebUIcaptureWebTestObject tool to detect content inside Shadow DOM, iframes, or Flutter Web elements.
  • [API Testing with AI] Improved UX in Generate Test tab:
    • Renamed Test Configuration to Configuration Setup
    • Added a dropdown menu to select existing test configuration
    • Added a dedicated New button lets you quickly clear all fields and start a fresh configuration
    Configuration setup in 11.1.2
  • Displayed test step description in HTML by default. If the test step has no description, test step code will display instead.
  • Added six new variables for Test Suite Collection email reports, allowing users to include total test case and test suite-level counts directly in the email:
    • totalTestCases
    • totalTestSuitesPassed
    • totalTestSuitesFailed
    • totalTestSuitesError
    • totalTestSuitesSkipped
    • totalTestSuitesIncomplete
  • Added support for editing variable default values in recorder workflows.
  • Added warnings on test steps when a locator matches multiple elements in Smart Mobile Recorder.
  • Added Chrome 147 and Edge 147 compatibilities.
  • [Security Compliance]:
    • Log4j minor upgrade.
    • Enforced TLS v1.2 to disallow insecure protocols.
    • Major upgrade of io.modelcontextprotocol.sdk:mcp-core — current supported MCP SDK version.

Fixes​

Fixes

Smart Mobile Recorder (SMR)

  • Fixed default dialog size appearing oversized with excessive whitespace.
  • Fixed overlapping text when highlighting test objects with errors.
  • Fixed function keys (Shift, Tab, etc.) not being sent correctly in Interactive mode.
  • Fixed elements not being clicked when a locator returned multiple matches in Interactive and manual mode.
  • Fixed Verify & Highlight throwing an error when using Custom locators.
  • Fixed inability to tap XCUIElementTypeStaticText elements inside a ListView.
  • Fixed variable type incorrectly changing to Method Call or flagging "Invalid test case" on save.
  • Fixed locator values auto-refilling after manual edits when navigating away and back.
  • Fixed iOS Predicate String locator generating incorrect values that failed to identify captured objects.
  • Fixed incorrect warning text displayed when renaming objects (was showing "Application ID" instead of the object name).
  • Fixed option verification checking only one locator instead of all selected locators.
  • Fixed the absence of an Undo option when removing a recorded or manually added step.
  • Fixed a deadlock that caused tests to always fail or hang on rerun when an unsaved test case was open.
  • Fixed Backspace and Enter keys not displaying correctly in the log view and during recording.

Recorder+ (R+)

  • Fixed navigateToUrl() steps being incorrectly captured when reloading the same page.
  • Fixed custom keyword UI not displaying parameters that have default values.
  • Fixed redundant findTestObject(null) being inserted into scripts after saving with an empty object field.
  • Fixed the Studio context menu being inaccessible within R+.
  • Fixed UI issues in the Add Variable screen.
  • Fixed error logs not displayed in red.
  • Fixed missing variable lists, incorrect list nesting for custom keyword parameters of type List, missing checkpoint selection, and invalid default value handling when editing parameters.
  • Fixed invalid custom desired capabilities not showing a proper validation error dialog before recording.
  • Fixed disabled steps being automatically re-enabled after saving.

SMR & R+

  • Fixed Call Test Case always showing a null ID regardless of the test case name entered, and when saving scripts.
  • Fixed inline action buttons of the next step overlapping an open context menu, blocking menu selection.
  • Fixed object rename synchronization so edited object names correctly propagate to scripts.
  • Fixed recording with a callTestCase step requiring users to save twice for the script to be saved correctly.
  • Fixed variable binding issues preventing users from calling variables into test steps and converting property-type variables correctly.

Web Recording Agent (Web RA)

  • Fixed collapsed prompt area disabling the AI Agent when clicking the Edit icon.
  • Fixed prompt area not enforcing a size limit when generating with a long prompt.
  • Fixed an intermittent error dialog appearing during AI script generation.

Log Viewer

  • Fixed missing execution information for tests that do not include a Start Browser or Start Application step.
  • Fixed duplicated nodes appearing in Log Viewer for test cases using variables and binding in test suites.
  • Fixed inconsistent execution status colors and progress icon state in the new execution UI.
  • Fixed Call Test Case input data not displaying correctly in the execution UI.

Execution & Job Progress

  • Fixed Incomplete status not appearing in the test suite tail in Job Progress.
  • Fixed the Job Progress Console Viewer window list not being cleaned up when Job Progress is cleared.
  • Fixed the Execution Information label in Test Suite being unable to expand, preventing users from setting retry options.
  • Fixed mobile environment labels displaying as constant text variables instead of their configured label names.
  • Fixed an error preventing test suite collections from opening in certain projects.

Self-healing

  • Fixed a regression where opening projects from older Studio versions reset Self-healing locator priority order for Web UI.
  • Fixed classic self-healing not being disabled at runtime when all locator methods were deselected.

Test Execution Cloud

  • Fixed canceling a cloud device connection not actually stopping the background process.
  • Fixed Test Execution Cloud failing to close the application after test execution.
  • Fixed execution failures for cloud mobile browser runs caused by unsupported mixed Selenium/Appium driver operations.

Katalon AI Assistant

  • Fixed intermittent exception "'Unauthorized, status code 401' when invoking REST Client method: 'com.katalon.testops.restclient.AdminServiceRestClient#getKseSubscription..." requiring a Studio restart to recover.
  • Fixed attachment failures when test case variables contained excessively large values, with error message "Object not found with path: studio/studio-assist/conversation-data/a-743630/o-743630/u-12715402/sa-chat-file-upload/8dc2a37b-5406-46f3-b232-4ae30c453a54".
  • Fixed a recurring "LlmApiRuntimeException: Cannot get Organization Id" error affecting multiple account types and versions.

General

  • Fixed missing console log steps for test cases using @Setup and @Teardown annotations.
  • Fixed intermittent Studio freezes when interacting with the Groovy script editor.
  • Fixed custom keywords not loading from the Plugins folder after a close-and-clean project operation.
  • Fixed Smart Locator parsing errors for locators containing escape characters.
  • Fixed the Web View UI becoming blocked when an arbitrary file was dropped into it.
  • Fixed missing platform global variable values when running a Test Suite Collection on Test Execution Local with Linux.
  • Fixed HTML report steps not appearing when the Split report into files option was enabled.

Windows Recorder

  • Fixed Windows Recorder failing to start due to an incorrect local fla server status check.
  • Fixed inability to drag and drop Windows objects in the object repository (Web and Mobile objects were unaffected).

Version 11.1.1​

Release date: April 7, 2026

Fixes​

  • Fixed an issue where users were unable to activate offline license in KS 11.1.0. To continue using offline license, download version 11.1.1.

Version 11.1.0​

Release date: April 1, 2026

This release focuses on AI and MCP enhancements alongside a product rebrand. Key new features include a standalone KS MCP Server mode, STDIO transport support for community MCP servers, built-in GitHub Copilot integration, GitHub MCP Server support, AI-powered test failure analysis via the analyzeTestFailure tool, and new CLI flags for fine-grained AI Self-healing control.

Breaking changes​

  • Renamed in-app UI product names:
    • Katalon TestOps to Katalon True Platform
    • Katalon Test Execution - Local to Katalon Test Execution - Local
    • Katalon TestCloud to Katalon Test Execution - Cloud
  • For Test Execution - Local command syntax, -ai.* is the new, supported syntax. -studioAssist.* arguments still work, but are deprecated. For users with scripts using -studioAssist.*, we recommend migrating to -ai.*.
  • Removed the older built‑in MCP tools getTestCases and getTestObjectsList and replaced them with the more general readFile and listEntityFiles tools.
  • The option for AI API Test Generation has been removed from the Import OpenAPI dialog.
  • Rebranded StudioAssist to Katalon AI Assistant and refreshed AI branding.

New features​

Use Studio MCP Server outside of Katalon Studio, Katalon AI Assistant MCP server configuration​

  • What's new: Added standalone KS MCP Server mode so you can run the KS MCP Server without opening the Katalon Studio IDE, authenticated via API key and governed by KSE licensing.
  • Why this matters: New KS MCP Server configuration panel lets you enable/disable the server, change port, configure authentication token, choose exposed tools, and copy connection config for external AI clients.
  • How it works: Refer to our guide on Configuring Katalon Studio MCP Server.

STDIO transport support for MCP servers​

  • What's new: Added STDIO transport support for MCP servers, enabling direct integration with community MCP servers that communicate over STDIO.
  • Why this matters: New transport type to allow broader MCP server compatibility, where you can integrate with many community MCP servers that use STDIO.

AI self-healing commands for Web and Mobile​

  • What's new: You can now control AI Self‑healing for Web and Mobile directly via Test Execution - Local command‑line.

    For Web AI Self‑healing:

    • -execution.selfheal.web.ai.enabled to enable/disable AI self‑healing for Web UI in that run.
    • -execution.selfheal.web.ai.inputSourcesto choose which data sources AI uses.

    For Mobile AI Self‑healing:

    • -execution.selfheal.mobile.ai.enabled to enable/disable AI self‑healing for Mobile.
    • -execution.selfheal.mobile.ai.inputSources to control mobile AI data sources.

  • Why this matters: You can explicitly enable/disable AI self‑healing and choose exactly which data sources (DOM, accessibility tree, screenshots) are used for more reliable, predictable, and tunable AI‑assisted test execution.

GitHub MCP server support​

Katalon AI Assistant now supports connecting to GitHub MCP Server via STDIO or HTTP with Personal Access Token (PAT) - so you can use GitHub‑hosted assets (e.g., existing automation tests) as AI context in Katalon Studio.

Note: HTTP with OAuth is not supported yet.

Troubleshoot with AI using analyzeTestFailure tool​

What's new: You can now trigger Troubleshoot with AI from Log Viewer/Report/Test Suite results with execution log attached directly to AI chat and for faster debugging. The AI failure analysis tool analyzeTestFailure is leveraged in Agent mode to analyze failed test cases/suites and suggest root causes and fixes within chat. How it works: Watch the demo video for a quick visual walkthrough of how this feature works.

Support custom file attachments in email reports​

Email report settings now allow specifying custom file extensions so user‑generated artifacts (e.g., .txt, .xlsx, .json) can be attached with reports.

Supported report formats in 11.1.0

Enhancements​

Enhancements

Katalon True Platform integration

  • Added a Connect to True Platform dialog to guide users through platform integration setup from within Katalon Studio.
  • Integration with Katalon True Platform and automatic test result upload are now turned on by default for new projects.
  • The login flow now opens in the system default browser, with a post-sign-up redirect to Katalon True Platform.
  • Updated the AI branding logo across Katalon Studio UI, including the main toolbar, chat panel, and AI failure analysis email reports.
  • Added a new Test Step to the Test Suite onboarding tour and introduced a Knowledge Hub banner.
  • Override authentication now includes an Account dropdown and an Org-scoped project list for more precise project targeting.

Katalon AI Assistant and MCP

  • Added shortcut to open the MCP configuration file in Studio for use cases like adding custom headers for an HTTP MCP server setup. You can also edit already-connected MCP servers from these settings.
  • Unified Studio built-in tools so Katalon AI Assistant can list and read arbitrary files in the project, not just predefined artifact types.
    • New tools added:
      • listEntityFiles to recursively list files and directories in the project
      • readFile to read the content of any file in the project, not just Katalon-specific artifact types
    • Tools removed: getTestCaseList, getTestObjectsList – replaced by listEntityFiles
  • The built-in MCP server's test case tools now support creating and updating test case variables. Supported variable types: String, Number, Boolean, Null.
  • SSE/HTTP MCP server connections now support configurable custom HTTP headers, enabling authentication and enterprise proxy scenarios.
  • Improved the Katalon AI Assistant settings UI/UX for Ask Mode, including layout and interaction adjustments for clarity.
  • Improved webUICaptureWebTestObject tool to display the list of created test objects in the chat once it completes.
  • Improved the clarity of the error message shown when Katalon AI Assistant reaches a request timeout.
  • Katalon AI Assistant now automatically reconnects to MCP servers when an authentication token expires during tool execution, without requiring a manual reconnect.

Recorders

  • Support editing recorded step parameters (string, number, null, boolean, global variables, test data, test objects, properties) directly from the UI.
  • Playback (Run All) in Smart Mobile Recorder now restarts the device session to ensure a clean starting state, with an optional Don’t show again confirmation dialog.
  • Improved naming to show cloud apps in a more readable app names along with identifier, version, and build for easier selection.
  • Smart Mobile Recorder now supports X,Y viewport coordinate capture on hover, enabling coordinate-based interactions to be recorded.
  • When a test object is selected via Screen Object in Smart Mobile Recorder, its viewport position is now shown alongside the highlight rectangle.
  • Mobile Recorder and Object Spy dropdowns now display the user-defined application name instead of the raw uploaded file name.
  • Traditional Self-Healing and AI Self-Healing configurations are now separated in Project Settings, allowing independent control of each mode.
  • Updated the label "Katalon Compact Utility" to "Katalon Studio Recording Engine packed extension" in Recorder Preferences to better reflect its function.

Other

  • The Expand All button in HTML reports now toggles the full execution tree, making it easier to review deeply nested test steps.
  • Added Chrome 146 and Edge WebDrivers 146 compatibilities.
  • Added notification dialog when an integration session becomes invalid, with a prompt to re-authenticate.
  • Removed the auto-apply logic for the default Appium version, giving users full manual control over Appium version selection.
  • Removed the redundant Org dropdown from the Override authentication panel.

Fixes​

Fixes

Web Recorder Plus

  • Users could not switch to another browser after playback when recording was done with a browser profile.
  • Recording with a browser profile incorrectly enabled both embedded and manually installed extensions at the same time.
  • An error dialog appeared on playback when a previous playback session had not fully finished.
  • Users were unable to select text using the mouse while editing a test step.
  • Recording with a Chrome profile failed if a previous recording session used Active Edge.
  • The "Switch to window title" step was missing when a new tab was automatically opened from a link or when switching between browser tabs during recording.
  • The legacy Web Recorder was unable to play back previously recorded test steps.
  • Existing objects were not reused when continuing a recording on an existing test case.
  • Recorder Plus recorded a redundant click() on the password field before setEncryptedText().
  • Recorder Plus re-created existing test objects instead of reusing them when continuing a recording.
  • Saved scripts were not updated with new steps when continuing a recording on an existing test case.
  • Continuing a recording with Clean Session incorrectly created a new test case instead of updating the existing one.
  • Recorder always used Chrome for playback regardless of which browser was used during recording.
  • Browser info displayed "Unknown" instead of the actual browser name in the recorder.
  • Recorder captured navigateToURL() instead of newTab() when a new tab was opened during recording.
  • The recorder showed a Delete button instead of Cancel when editing a recording step.
  • Recorder did not recognize Global Variables when recording was done with a non-default execution profile.
  • Web Recorder with Edge Profile captured the URL of the default window instead of the one being recorded.
  • Recorder captured the ENTER key as a string value instead of a key action.
  • The Start Record section did not auto-collapse when using Active Browser mode.
  • Recorder was unable to recognize ENTER key input in a text field.
  • Active Browser playback launched a different browser than the one used during recording.
  • Recording Engine Settings were not saved correctly when toggled.
  • Smart Locator still captured the locator value even after being unchecked.
  • No "Profile Not Found" warning was shown when starting a recording with a Chrome profile that no longer exists.

Smart Mobile Recorder

  • Execution status was shown as Failed when the user stopped a re-run script manually.
  • Smart Mobile Recorder did not check whether Android SDK was installed before starting a recording session.
  • Both SMR dialogs could not be dragged independently when using dual monitors.
  • Switch to Landscape/Portrait did not work on iOS and Android devices.
  • Objects saved as "Duplicate" were stored correctly but the generated script referenced the wrong object name.
  • The device dialog position reset to default on every open regardless of where the user had previously moved it.

Katalon Test Execution - Local & Test Execution

  • Katalon Test Execution - Local failed immediately after plugins loaded due to an invalid byte exception.
  • API tests failed on Hyper-execute after upgrading Katalon Test Execution - Local to v11.0.1.
  • Existing object locators failed after upgrading from Katalon Studio 10.4.3.
  • Cucumber scripts failed to run inside a Test Suite Collection when executed from Katalon Test Execution - Local on TestCloud.
  • Katalon Test Execution - Local threw an "Invalid AI service configuration" error on startup.
  • The browserType value could not be overridden from Katalon Test Execution - Local when running a Test Suite Collection.
  • A false Chrome DevTools Protocol (CDP) version mismatch warning appeared in the Console on every test execution when using newer Chrome versions, even though execution was not affected.

Other

  • Fixed an issue where users were unable to add multiple drivers in Custom Desired Capabilities — instead the new driver replaces the previously selected driver.
  • [Katalon Tracer] Tracer view did not work when a test case called closeBrowser at the start.
  • Video recordings generated a file path that exceeded the maximum allowed length.
  • Auto-link to release did not work when using the new execution ID format.
  • HTML reports were missing test steps when the Split the report into files setting was enabled.
  • Fixed an error where Windows Object Spy failed when an element’s name contained the special hex character 0x1B, causing the WinAppDriver/Appium driver to throw invalid character errors.
  • Web Spy lost object data when saving with the Merge into existing objects or Replace existing objects option.
  • Some test objects captured during mobile recording were incorrectly shown as Web Objects.
  • Changing the browser in a Test Suite Collection also reset the execution profile to Default.
  • Desired Capabilities were automatically removed when clicking on other fields in a Test Suite Collection.
  • Auto Tagging did not work on the first launch of Katalon Studio.
  • Test suite names were incorrectly displayed in red in the Job Progress view.
  • Fixed garbled Chinese annotations in Script editor.
  • An intermittent "Unable to resolve class" and invalid package binding error occurred during test execution on cloud environments.

Version 11.0.1​

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: WebUIcaptureWebTestObject for generating Katalon-standard test object locators, and webUITakeScreenshot for capturing and displaying web page screenshots inline in Katalon AI Assistant. It also includes fixes across Smart Mobile Recorder, Web Recorder Plus, execution engine, Log Viewer, HTML reports, and Katalon Test Execution - Local.

New features​

  • Added new MCP Server tools for Katalon AI Assistant:
    • WebUIcaptureWebTestObject 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 Test Execution - Local **:

    • Katalon Test Execution - Local 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 Katalon Test Execution - Local .
    • [macOS] Katalon Test Execution - Local 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.

  • [Katalon AI Assistant]:

    • 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, Katalon AI Assistant Agent profile, and new MCP tools. It also includes UI enhancements and extensive fixes across Katalon AI Assistant, 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.
  • [Katalon AI Assistant | 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 Katalon AI Assistant 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.
  • [Katalon AI Assistant] 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 Katalon AI Assistant.
  • [UI] Re-organized components inside Katalon AI Assistant 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.
  • [Katalon AI Assistant]:
    • 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 Katalon Test Execution - Local .
  • 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?

Table of Contents