Claude Certified Architect – Foundations Practice Question 67: Prompt Style Matches Output

Claude Certified Architect, Foundations · Test 7 · question 7 of 10

Prompt Engineering & Structured Outputmedium

A customer-facing chat widget cannot render markdown. The system prompt is itself written with "##" headers, nested bullet lists and bold text, and includes the line "Do not use markdown." Claude still returns bulleted, bold-heavy answers. Which change is the best next step, assuming the model is a current Claude model?

Select one answer
Show answer and explanation

Answer:

  • D. Rewrite the prompt itself as plain prose, matching the style wanted in the output

D is correct. The best-practices guide notes that the formatting style used in your prompt may influence Claude's response style, and suggests matching your prompt style to the desired output style; for example, removing markdown from your prompt can reduce the volume of markdown in the output. Keep a clear instruction about what format you do want, written in the same plain style.

A is wrong because the Claude API rejects assistant prefill on Claude 4.6 and later models (it returns a 400 error), so it is not a dependable technique here. B is wrong because effort controls how much reasoning and work the model puts in, not how a conflicting prompt style is resolved. C is wrong because shouting a negative instruction inside a markdown-heavy prompt keeps the conflicting signal; the guide favors matching style, and stating what to do rather than what not to do.

Reference: Prompting best practices - Control the format of responses

Practice Test 7 ›

Study guides for Claude Certified Architect, Foundations