[vanpelt]

Tailwind is especially well suited for LLM generation because all of the styles come inline with the HTML. Could definitely just ask it to write styles but this approach also improves token efficiency.