Research questionHow can LLMs generate correct routines when correctness depends on their joint runtime behavior?Textual prompts may describe individual routines without determining whether their combined runtime behavior satisfies the task. As a result, code generation can fail when one routine’s meaning depends on another’s execution.