Hey, I made Koòrd FlatSkills to make working with LLMs a bit more pragmatic and productive. They’re open source, so if you feel like trying them, have a look. I’d be really interested in hearing how they behave in actual use, especially if you manage to break them.