FAQ

  1. What Generative AI provider do you use?

    We use OpenAI. You can find their privacy documentation here.

  2. Which cloud provider is used to provision Virtual Machines?

    Reflect uses Amazon Web Services (AWS) for cloud hosting, based in AWS's us-east-1 data center in Northern Virginia. Amazon Cloudfront serves as our CDN provider, with edge-caching of static assets globally. We ensure security by encrypting data both at rest and in transit. AWS's us-east-1 data center complies with ISO27001, SOC1, and SOC2 standards. Supporting materials are available upon request.

  3. How is login info securely sent to the cloud?

    We encrypt data from Zephyr Scale to Reflect servers and then into Reflect cloud browsers. Data remains encrypted at rest within databases. For entering passwords during test runtime, we've developed a "secrets manager" feature. This feature allows you to reference sensitive values by name, injecting them into tests at runtime. Learn more about using this feature here.

  4. How is PII data handled?

    We recommend using synthetic test data and testing against staging environments. While all data is encrypted at rest and in transit, we do not implement special measures for handling PII data.

  5. What applications can I test with this new capability?

    You can test various web applications. If you need mobile testing capabilities, please let us know.

  6. Can we test apps behind a firewall or in a private environment?

    Yes, you can use the Reflect agent for this. Find detailed instructions here. Once completed and with a Reflect account, we'll enable this feature for you.

  7. Can we use Reflect independent of Zephyr Scale?

    Yes, Reflect is a standalone SaaS platform for low-code functional testing.

  8. What types of AI prompts do you support?

    • For actions: click, hover, input, [Enter] keypress, navigate.

    • For validations: ask true/false questions about the page. For example, "Validate that the price is $2.10." We assume you're seeking a true response, and the test step retries until true or fails at the step timeout (default: 10 seconds).

    • We recommend making prompts as explicit as possible for optimal results.

  9. Will the feature be available for Zephyr Squad?

    No. The feature will be only for Zephyr Scale.

  10. Do you have an example prompt?

    Here is a sample manual test case validating the flow of adding items to the cart on the Smartbear’s e-comm website.  Use this link as a starting URL for you test case: https://bearstore-testsite.smartbear.com/

    1. Click on Sports on the navigation bar

    2. Select the Price filter value for ‘up to $25.00’

    3. Click on a Product

      • TEST DATA:

        • Titleist Pro V1x

      • EXPECTED RESULT:

        • Validate that the price is $2.10

    4. Update quantity to 10

    5. Add to Cart

    6. Click Checkout button

In this section:
© 2024
Publication date: