Test coverage is somewhat spotty right now. Plenty of the core pieces are covered, along with actual bugs that were discovered by usage, but there are plenty of obvious gaps in our current verification.

Any contribution to increase coverage of the pytest tests is welcome.

Note that these tests do not have a usablepy command, and so need to be structured to not rely on that - plenty of mocks, monkey-patching and directly importing commands is intended!

(For tests that can only be run after the full app is compiled/installed, create a new issue describing the scenario and why you think it wouldn't be valid as a unit test.)