CCAR-P Advanced Agentic Architecture Practice Question
Which of the following describes the role of the 'System Prompt' in an agentic architecture?
⚠ Common exam trap
Candidates frequently confuse the system prompt with dynamic few-shot learning examples, mistakenly thinking it handles turn-by-turn conversation memory rather than persistent global boundaries and agent identity.
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
✓
A mechanism for defining the agent's identity and constraints.
The system prompt serves as the 'DNA' of the agent, dictating its core behavior, constraints, and operational goals. By setting the context outside of the user's conversation stream, it ensures that the model adheres to defined roles and ethical boundaries consistently, regardless of user input. It provides the necessary structure that turns a raw LLM into a purposeful, reliable assistant within a larger system architecture.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
A temporary buffer for user-provided instructions.
Why it's wrong here
User-provided instructions go into the user message buffer, not the system prompt. The system prompt is specifically for architect-defined behaviors and instructions. Mixing user intent with system-level constraints in the same buffer makes the system highly vulnerable to prompt injection attacks.
- ✓
A mechanism for defining the agent's identity and constraints.
Why this is correct
The system prompt is the primary location for defining 'who' the agent is and the boundaries it must respect. This is an essential architectural component that dictates how the agent processes information, adheres to safety protocols, and interacts with tools, forming the foundation of its autonomous behavior.
- ✗
An automated script for executing Python code.
Why it's wrong here
System prompts are text-based instructions, not executable code. While they can tell an agent *how* to write or use Python code, they do not have the capability to execute scripts themselves. Execution is handled by a separate, secure environment where tools are invoked by the model.
- ✗
A method to store long-term user history.
Why it's wrong here
Long-term history is stored in an external database, not in the system prompt. Storing history in the system prompt would quickly exceed context window limits and become unmanageable. The system prompt is for static instructions, while conversation history is dynamic and managed through separate storage mechanisms.
About these practice questions
One of 262 original CCAR-P 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 CCAR-P 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 CCAR-P exam.