CCDV-F Prompt and Context Engineering Practice Question
A developer is using Claude to review a 180-page contract. The model must cite the exact page number for each risk it flags. Placing the entire contract in the prompt causes the model to cite vaguely or omit page numbers. Which change best improves citation accuracy without exceeding the context window?
⚠ Common exam trap
The trap here is believing that reducing temperature fixes attribution errors, when the real problem is that the page boundary is not represented in the text the model actually reads.
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
✓
Insert explicit page-marker tags such as '[[PAGE 42]]' before each page's text and instruct the model to cite the nearest marker.
Embedding page markers makes citation a local pattern-matching task instead of a global spatial inference. The model can still read the whole contract in one pass, preserving cross-reference reasoning, while the nearest marker gives an unambiguous page label. Temperature changes and summary-then-cite pipelines do not supply the missing positional signal.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Ask the model to summarize each page first, then cite the summary.
Why it's wrong here
Summarizing first loses the exact page position of each risk because the summary merges content across pages. The model would then be citing its own summary rather than the source, which is exactly the attribution failure the developer is trying to fix. It also adds an extra generation pass and more tokens.
- ✗
Lower the temperature to 0 so the model copies page numbers verbatim.
Why it's wrong here
Temperature affects sampling randomness, not the model's ability to locate and attribute information across a long document. At temperature 0 the output becomes more deterministic, but the model can still miss the page association if the page marker is not explicitly embedded in the text it reads.
- ✗
Split the contract into 180 separate API calls, one per page.
Why it's wrong here
One call per page destroys cross-page reasoning, which is often where contract risk lives, such as a defined term used 40 pages later. It also multiplies latency and cost by roughly 180 times. Page-level chunking can help retrieval, but it is the wrong tool when the task requires a global view of a single document.
- ✓
Insert explicit page-marker tags such as '[[PAGE 42]]' before each page's text and instruct the model to cite the nearest marker.
Why this is correct
Adding unambiguous page markers turns a spatial memory problem into a simple nearest-marker lookup. The model does not have to infer pagination from layout; it reads the marker adjacent to the text. This keeps the full document in context while making attribution a mechanical operation the model can perform reliably.
About these practice questions
One of 257 original CCDV-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 CCDV-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 CCDV-F exam.