CCAO-F Prompting and Context Engineering Practice Question
A financial analyst uses Claude to summarize a 60-page quarterly earnings report. The prompt includes the full report between <document> tags and asks for a 200-word summary of key risks. Claude's summary frequently omits risks mentioned in the middle of the report. What is the most effective change to the prompt to improve recall of mid-document content?
⚠ Common exam trap
The trap here is assuming that a single long-context call will uniformly attend to all parts of a large document.
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
✓
Split the report into smaller sections, summarize each section separately, then combine the summaries in a final pass.
Long documents can suffer from the lost-in-the-middle effect, where content in the center receives less attention. Splitting the document into smaller sections and summarizing each ensures that every part is processed with sufficient focus. A final aggregation step combines the section summaries, preserving mid-document risks that would otherwise be missed.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Move the instruction and the risk summary request to the end of the prompt, after the document.
Why it's wrong here
Placing instructions after the document can help with instruction following, but it does not address the recall problem caused by long-context attention distribution. The model may still underweight middle sections. This change alone is insufficient because the core issue is how the document is presented and chunked, not where the instruction sits.
- ✗
Lower the temperature to 0 to make Claude more deterministic and factual.
Why it's wrong here
Temperature affects randomness in generation, not attention over long inputs. A deterministic setting will not recover risks the model failed to attend to. While lower temperature can reduce creative drift, it does not solve the structural recall issue in long-context summarization.
- ✓
Split the report into smaller sections, summarize each section separately, then combine the summaries in a final pass.
Why this is correct
Chunking the document into smaller sections and summarizing each reduces the context length per call, mitigating the lost-in-the-middle effect. A final aggregation pass ensures no section is skipped. This directly targets mid-document recall and is the most reliable fix for the described symptom.
- ✗
Increase the max_tokens parameter to allow a longer summary that can include more risks.
Why it's wrong here
Max tokens controls output length, not the model's ability to attend to or recall input content. A longer summary does not guarantee that middle sections are read. This parameter change addresses output truncation, which is not the reported problem.
About these practice questions
One of 259 original CCAO-F practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. 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.