A test-generation reviewer is inconsistent about whether an untested error branch counts as a coverage gap worth reporting. The team wants consistent judgment that also carries over to cases it hasn't seen. Which additions to the prompt best achieve this? (Select two)

Answer: A and B — Two or three borderline cases, each explaining why it was or wasn't reported / A contrast pair: an error branch exercised indirectly by another test, and one never exercised

A few targeted examples of ambiguous cases with their reasoning (a), and pairs that separate acceptable patterns from genuine gaps (b), teach judgment the model can generalize. Clear-cut examples (c) cover cases that already go right. An exhaustive list (d) only matches pre-specified cases and doesn't generalize to new code.