During the software development life cycle, there is rarely a case when you want to run all existing tests in your Karate project; for example, when a full regression test is necessary on your last environment prior to production deployment. Most of the time, it is sufficient to only run a smaller set of tests that fit the current use case.

Examples of this could be as follows:

• During feature development, only the tests associated with the feature under development are run
• During test development, only the current test under development is run to make sure it tests what it should
• After deployment, a few smoke-test scenarios are run to verify that it was successful
• A small set of tests covering the most important flows are run against the live instances of APIs for monitoring purposes

This is of course completely dependent on your or your team’s requirements.

To make it easier to show this, we created a new demo project...