Technical Discussion

Hi community,I am using the Standard Module Check Webpage Accessibility (RunHtmlAccessibilityAnalysis) to scan multiple pages.When exporting the accessibility report JSON file, the summary block shows total numbers for Passed and Failed. However, under the screens list, only violations are listed for each page. The detailed rules that Passed or were Inapplicable are missing.For our audit evaluation, we need visibility into which specific Axe-Core rules passed on each page to determine our manual testing delta. Is there a way to force Tosca to include detailed passes and inapplicable rule arrays inside each screen object? If this is by design, what is the best practice in Tosca to retrieve the complete Axe-Core result object? Thanks in advance!

I am using a Shift Sync account with my organisation's email address, and I am leaving this organisation now. How should I change my Email address for my Shift Sync account?

Everyone is excited that AI can write tests.Far fewer teams are asking a more important question:Who is testing the prompts?In traditional software, business logic lives in code.In AI applications, part of the business logic lives in prompts.That means a prompt change can introduce a bug just as easily as a code change.I've seen teams:• Change a single sentence and break production behavior.• Improve one use case while degrading five others.• Fix one prompt and accidentally increase hallucinations.Prompt engineering isn't just about writing better prompts.It's about building prompts that are:✅ Testable✅ Repeatable✅ Versioned✅ MeasurableIf your prompts change but you have no regression suite…You're deploying blind.As AI applications mature, prompt testing will become as normal as unit testing.The teams that start today will have a significant advantage.How does your team validate prompt changes before they reach production?#AI #PromptEngineering #SoftwareTesting #QualityEngineering #LLM

Hello all,we are integrating our Tosca Cloud tenant with SAP Cloud ALM (Test Automation Endpoint).For this integration we need an API client with OAuth2 Client Credentials (Client ID & Client Secret) with the permission "Test Automation". We have checked both "System settings" and "Integrations" in our Tosca Cloud tenant, but there is no option available to create an API client ourselves. We have Tenant Admin rights, but this does not seem to be sufficient to generate OAuth2 credentials via self-service.  

Hello,I am trying to figure out if this is even possible.  I want to have for ‘one swagger api’  an API test module that is ‘generic’ that I can feed the method and all the various request items to it and send using the TCD test sheet. So think of it as one module but ‘many tests’ including type, test data, methods different (POST/GET) etc.   Is this a possibility? I have tried but seem to be limited to just the ‘one’ URL  Ideas anyone? We can’t use external tools as we are on premisis and secure but want the maximum flexibility without having to do rescans every time or have a separate module for each type of call?  ThanksI have one test of one endpoint with a bunch of security tests I can run in Test sheet varying the values of fields like token, user, parameter data, etc but that seems to be all I can do.appreciate any help,

Give it a try. More to come  https://www.youtube.com/shorts/5Eay25HSuzw

Hi Team,We are planning to conduct a PoC for a Salesforce project. We would like to understand Tosca’s capabilities for Salesforce automation, including its support for rapid automation. Can anyone share your experiences with us.Thanks,Sudha

Hey all!I would like to buffer a regex value. I have a very long payload with customer data and I’d like to extract only the customer’s number. This whole payload is stored in “Result”. I have built this regex: {REGEX["customerNumber":"[0-9]+"]}And I want to store this extraction into a buffer.Anyone can support please? Thank you!

Please see below steps taken in the screenshot.I am getting below error message. Please advise. 

Hi everyone, I'm currently working on integrating Generative AI into our test automation framework and could use some expert advice. Has anyone had experience using AI to enhance test case generation or maintenance? Any recommendations on best practices or tools would be greatly appreciated!

Fiori application scanning of a drop down used to show the element type as ‘Dropdown’ before but now it is showing as ‘Input’ like a text field. We can scan the element, but the drop-down options are no longer showing for selection in TOSCA (which they were used to properly show before). This is blocking the scripting as entering a text in Dropdown is not possible.

For our new product, I have the Figma design and not having requirement document. In this figma design, I have more detailing. How can I convert the design to test scenarios and script in a easy way and also how we can automate the same since manual effort is huge for this

Greetings,Hope having a fantastic December holidays to you all!Allow me to request you guys’ help for the following issue!After running the DEX execution overnight, some testcases are updated with the updated execution logs while some are not updated at all.Looking forward to you guys’ suggestions!With Regards,Zuzan   

Hi All,Has anyone integrated qTest with Selenium?We’re looking to trigger test execution from qTest and automatically return the results back into qTest.Any guidance or shared experiences would be greatly appreciated.Cheers, Jag

The Screenshot captured by 2024.2 version of Tosca captures partial screen despite the setting for Screen set to everything. Could you please guide me with the issue here? Full screen view:Full screen viewTosca Screenshot:Tosca Screenshot 

This is the complete error i am getting while trying to scan- Impossible to connect to Appium Server at 'http://127.0.0.1:8585/'. Please, check if the address is correct and connection is possible. Original error: 'Command 'C:\\Users\\rishi_harjani\\AppData\\Local\\Android\\Sdk\\platform-tools\\adb.exe -P 5037 -s 9b09478b shell 'settings delete global hidden_api_policy_pre_p_apps;settings delete global hidden_api_policy_p_apps;settings delete global hidden_api_policy'' exited with code 255'; Command output:  Also, This is the article(KB0015473) i followed to setup environment variable but nothing works. same issue. My android real device is successfully connected and visible in TMA but while trying to scan , I am getting the error. Any idea/suggestion how to resolve this. ?

Traditional Intel core I5/I7 laptops are very slow when running LLM models locally. These laptops does not have GPU. hence slow performance is expected. Hence, planning to upgrade to a new Windows laptop where I could run LLM models with satisfactory performance. Again ultra high-end models would be very expensive.Can I please ask community members to share their experience and recommendations for Windows laptop  

Hello, has anyone experiance with the automation of an application created with teh Vaadin framework?https://vaadin.com/ We are struggeling with the identification of the gui objects. Thanks,Volker

The error wat am getting whole triggering the events please make sure that you are authorised to access the service endpoint https//tricentis.host:443/automationobjectservice/Events. This is the error am receiving. I have checked the dex setup , authenticated the api tokens and have given the access tokens to the agent restarted the dex execution services checked the server details inside the commander all the Authorization everything is correct and inside the agent machine the communication is also working. But still this error is triggering .  Can I get a step by step solution why this error is triggering. Even though the SSL certificates everything is correct 

We have a scenario with two dropdowns. Based on the selections made in both dropdowns, the third field is sometimes enabled or disabled. Instead of using an IF statement with a condition, we attempted to pass the condition directly. For the third field: We entered the value as Enabled -> <name> and set the Action Mode to Buffer. For the same third field (copied and pasted), we passed the condition as {B[name]} == "TRUE" and entered the text value in the Test Step Value column with the Action Mode set to Input. 

Do you have to use Claude Desktop with NeoLoad or is there another way to use their MCP? Can you use web versions of Claude or another chat tool?

Hi Team,Try to request your help to check my license issue, I already got below mail for the license request but the link cannot work once click will show error ‘Docs.tricentis.com reject to connect’ , can anyone tell me what’s the correct setup for the web browers or what should I do?  I already set the security of browser as ‘lowest’. thanks so much.