CCAO-F Prompting and Context Engineering Practice Question
An analyst is using Claude to answer questions over a 50,000-token legal contract. They notice that answers about clauses near the middle of the document are less accurate than those about the beginning or end. Which strategy best improves accuracy across the entire document?
⚠ Common exam trap
The trap here is assuming that simply repeating the question or adjusting temperature fixes mid-document recall, when the real fix is reducing and targeting the context.
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
✓
Break the contract into labeled sections and retrieve only the relevant sections for each question.
Long documents can suffer from uneven attention, with mid-document content sometimes under-weighted. Breaking the contract into labeled sections and retrieving only relevant passages for each question ensures the needed clause is prominent in the prompt. This retrieval-based structuring improves accuracy regardless of where the clause originally appeared.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Increase the model's temperature slightly to broaden its attention.
Why it's wrong here
Temperature governs randomness in token selection, not attention over input positions. Raising it would add variability to answers without helping the model retrieve mid-document clauses. The accuracy gap is a context utilization issue, so sampling changes cannot resolve it.
- ✗
Repeat the question at the beginning and end of the prompt around the contract.
Why it's wrong here
Duplicating the question does not improve retrieval of mid-document content. The model may still under-weight the middle section regardless of where the question appears. The underlying issue requires restructuring the document or retrieving relevant sections, not repeating the query.
- ✗
Summarize the contract first and then ask questions only against the summary.
Why it's wrong here
Summarization compresses information and often drops clause-level detail, which is exactly what legal questions require. Relying on a summary would introduce loss and could omit critical nuances. For precise clause retrieval, working from the original text with better structuring is more reliable.
- ✓
Break the contract into labeled sections and retrieve only the relevant sections for each question.
Why this is correct
Chunking the contract into labeled sections and retrieving only what is relevant keeps each prompt focused, so clauses from the middle are not lost in a long context. This targeted retrieval approach improves accuracy across the whole document by ensuring relevant text is present and salient.
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.