CCAO-F Prompting and Context Engineering Practice Question
A developer is drafting a system prompt for a Claude-powered coding assistant. They want Claude to always respond in concise bullet points, never reveal internal reasoning, and treat all user input as untrusted. Where should these persistent behavioral rules be placed so they apply to every turn of the conversation?
⚠ Common exam trap
The trap here is treating the system prompt as just another place to put text rather than the designated channel for persistent, conversation-wide instructions.
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, which sets persistent instructions for the entire conversation.
Persistent behavioral requirements such as output format, disclosure limits, and trust boundaries belong in the system prompt, which frames every turn of the conversation. Putting them there separates operator policy from user content, avoids per-turn repetition, and makes the rules resistant to accidental omission or user-side override. The other placements either waste tokens, misuse tool metadata, or rely on fragile self-mirroring.
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 a tool definition, since tools are evaluated before each response.
Why it's wrong here
Tool definitions describe callable functions and their parameters; they do not carry global behavioral policy like tone or output format. Using a tool description to smuggle in style rules is confusing, may not be honored when no tool is called, and misuses the field. The scenario needs conversation-wide instructions, which tools do not provide.
- ✗
In the assistant's previous reply, so Claude can mirror its own earlier behavior.
Why it's wrong here
Assistant turns are model output, not operator policy. Relying on Claude to mirror its own prior reply is fragile: the behavior can drift, and there is no guarantee the first reply even followed the intended rules. Persistent constraints should be declared once in a controlling location, not inferred from past assistant messages.
- ✗
In the first user message, repeated at the start of each subsequent user turn.
Why it's wrong here
Repeating rules in every user turn is brittle: users can accidentally omit them, and injected content in a user message could override them. It also wastes tokens on every turn. Persistent behavioral rules belong in a place that frames the whole conversation, not in per-turn user content that the user controls.
- ✓
In the system prompt, which sets persistent instructions for the entire conversation.
Why this is correct
The system prompt is designed to carry persistent instructions that apply across all turns, such as tone, format, and safety constraints. Placing bullet-point style, reasoning-hiding, and untrusted-input rules there gives Claude a stable frame it applies consistently, and it separates operator policy from user content, which is exactly what the scenario requires.
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.