FAQ:

Testhelper is designed to facilitate test automation and solve common testing problems, so:

• It is recommended to use it in a test Docker environment: raise a project, services (databases, brokers, etc.) in docker containers in the GitLab (or analogues) pipeline or locally (more for debugging and writing autotests), when using real services, problems are possible (the tool will not meet the timeout of responses, consumers, etc.).
  

• Testhelper designed to run tests in one thread, which is due to both stability and follows the principle “clean everything before each test”, of course, you don’t have to resort to cleaning and you can use generated data, but stability is not guaranteed when running tests in multiple threads. If you have a lot of tests and their execution time is longer than acceptable, it is recommended to parallel the execution of different jobs in the pipeline (this ensures complete isolation).
  

Yes. It is highly desirable to share your experience and suggestions.

Definitely stability.

Yes. Roughly speaking, the application from the outside works the same in any language. But this does not coincide slightly with the standard (for me) principle of test automation: autotest language = application language.

Indeed, I was inspired by the codeception with whom I loved working on one of the projects.

Like everything in Java it should, but it was systematically used on Linux (locally and in docker). If something doesn’t work, we’ll fix it (the methods that called grep have been removed).

No, testhelper is designed for backend testing. For frontend I recommend using playwright or something similar..