CCAO-F Prompting and Context Engineering Practice Question
A developer is using the Messages API and wants Claude to adopt a strict persona as a compliance reviewer for every request in a long conversation, without repeating the persona instructions in each user turn. The persona must apply to all messages and should take precedence over casual instructions the user might add later. Where should the developer place these persona instructions?
⚠ Common exam trap
The trap here is assuming that any early user message is equivalent to a system prompt, when only the dedicated system field provides persistent, higher-priority framing across all turns.
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
✓
In the system prompt parameter, which is applied across all turns of the conversation.
The system prompt is the correct location for instructions that must frame every turn and carry higher priority than user turns. It persists across the conversation without being repeated, and Claude is trained to respect system-level guidance over casual user instructions, which satisfies the compliance persona requirement.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
In the system prompt parameter, which is applied across all turns of the conversation.
Why this is correct
The system prompt is a dedicated field that frames the entire interaction and is not part of the user-visible transcript, so it consistently applies to every turn. Claude is trained to treat system-level instructions as higher-priority guidance, which matches the requirement that the compliance persona take precedence over casual later user instructions.
- ✗
As a trailing assistant message that restates the persona before each user reply.
Why it's wrong here
Assistant turns represent the model's own prior output, so injecting persona text there fabricates a model statement rather than instructing it. It also bloats the transcript and can be overridden by later user turns, making it unreliable for a persistent, high-priority persona across an entire conversation.
- ✗
In the first user message of the conversation, before the actual task.
Why it's wrong here
Placing the persona in the first user message works for a single turn, but in a multi-turn conversation later user messages can dilute or contradict it, and the model may weight the most recent user instruction more heavily. It also requires the developer to preserve that message across every call, which is fragile compared to a dedicated system field.
- ✗
In the metadata field attached to the request for logging and traceability.
Why it's wrong here
Metadata is intended for tagging and observability, not for influencing model behavior. Claude does not treat metadata as instructions, so persona text placed there would have no effect on the generated response and would simply be stored alongside the request for analytics purposes.
About these practice questions
Courseiva writes every CCAO-F question from scratch — 259 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or 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.