CCAO-F Prompting and Context Engineering Practice Question
A legal-tech startup uses Claude to summarize deposition transcripts that are frequently 80,000 to 120,000 tokens long. Early tests show the model sometimes ignores instructions placed near the top of the prompt and produces summaries that omit late sections of the transcript. Which TWO prompt-engineering changes should the developer make to improve instruction adherence across the full context? (Choose two.)
⚠ Common exam trap
The trap here is treating a long-context attention problem as a generation-parameter problem, and reaching for temperature or max_tokens when the real levers are instruction placement and document structure.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Wrap each deposition section in clearly labeled tags such as <section id="12"> and require the summary to cite section IDs.
Long-context adherence improves when the instruction sits after the material it governs and when the document carries structural anchors the model can traverse. Placing instructions last exploits stronger attention near the end of the prompt, while labeled sections with a citation requirement force coverage of every part of the transcript and make omissions detectable.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Repeat the summarization instructions verbatim in the system prompt and again immediately before the transcript.
Why it's wrong here
Duplicating instructions does not fix positional dilution, and placing a copy before a very long transcript recreates the same problem. The repetition also consumes context budget that could hold transcript content. The issue is where the instructions sit relative to the material, not how many times they appear in the prompt.
- ✗
Raise the temperature to encourage the model to explore more of the transcript.
Why it's wrong here
Temperature affects sampling randomness in the generated output, not which parts of the input receive attention. Higher temperature would make summaries less consistent and more prone to embellishment, which is especially harmful in legal summarization. It does nothing to counteract the positional dilution causing late sections to be dropped.
- ✓
Wrap each deposition section in clearly labeled tags such as <section id="12"> and require the summary to cite section IDs.
Why this is correct
Structural tags give the model navigable anchors throughout a long document, and requiring section citations forces it to traverse the entire transcript rather than summarizing only the opening. The citation requirement also produces verifiable output, since a missing section ID is immediately visible during review.
- ✗
Increase the max_tokens parameter so the summary can be longer.
Why it's wrong here
Output length has no bearing on whether the model attends to late sections of the input. A longer summary can still neglect the final third of the transcript. The reported failure is about input attention and instruction placement, so raising output limits addresses a different constraint entirely.
- ✓
Place the transcript content first and the summarization instructions after it, near the end of the prompt.
Why this is correct
Long-context performance improves when instructions follow the material they govern, because the model's attention near the end of the prompt is strongest. Putting the transcript first and the instructions last keeps the task directives close to the point of generation, reducing the chance that early instructions are diluted by tens of thousands of intervening tokens.
About these practice questions
This CCAO-F question is part of Courseiva's 259-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Anthropic exam blueprint
This CCAO-F practice question is part of Courseiva's free Anthropic certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the CCAO-F exam.