Claude Certified Architect – Foundations Practice Question 77: Explicit Scope Newer Models
Claude Certified Architect, Foundations · Test 8 · question 7 of 10
After migrating to a newer Claude model, a team notices that the prompt "Create an analytics dashboard" now produces a correct but bare-bones result, where the previous model added filters, charts and interactions unprompted. They want the richer output back. What is the recommended fix?
Show answer and explanation
Answer:
Anthropic's prompting guidance says Claude responds well to clear, explicit instructions, and that if you want "above and beyond" behavior you should request it explicitly rather than rely on the model to infer it from a vague prompt. Adding modifiers that ask for more features and detail shapes the output; the documented example uses exactly the wording in B.
- A is wrong: max_tokens is only a ceiling on length; the output was short because the request did not ask for more, not because it was truncated. - C is wrong: shouting or repeating the same vague request adds no new information about the desired scope. - D is wrong: prefilling does not request additional features, and assistant-turn prefill is not supported on the newest models.
Reference: Prompting best practices