In my experience they are great for test tooling. For actual tests after I have covered a number of cases it's very workable to tell it to identify gaps and edge cases and propose tests than I'd say I accept about 70% of it suggestions.
While people's experience with LLMs is pretty varied and subjective, saying they're bad at writing tests just isn't true. Claude Code is incredible at writing tests and testing infrastructure.
It worth mentioning that one should tell CC to not overmock, and to produce only truly relevant tests. I use an agent that I invoke to spot this stuff, because I've run into some truly awful overmocked non-tests before.
