CCAR-P Advanced Agentic Architecture Practice Question
An architect is designing an agent to perform administrative tasks in a corporate environment. One of the tools allows the agent to delete employee records. What is the most important architectural safeguard to implement for this specific tool?
⚠ Common exam trap
Candidates often choose automated validation or logging as the primary safeguard. They underestimate the risk of model hallucination in high-stakes actions, assuming that code-level checks are sufficient to prevent irreversible business data loss.
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
✓
Human-in-the-Loop (HITL) approval step
In high-stakes agentic environments, Human-in-the-Loop (HITL) is the primary safety mechanism for irreversible actions. While automated checks are helpful, human oversight ensures that the agent's reasoning aligns with organizational policy and prevents catastrophic data loss from potential hallucinations or misinterpretations of user intent in sensitive contexts.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Strict regex validation on the employee ID
Why it's wrong here
Regex validation only ensures that the format of the identifier is correct, not that the deletion itself is authorized or appropriate. An agent could provide a perfectly formatted ID for the wrong employee, leading to an unauthorized deletion that regex validation would be completely unable to detect or prevent.
- ✗
Using a smaller, more focused model for deletion
Why it's wrong here
A smaller model is generally less capable of complex reasoning and more prone to errors than a larger model like Claude 3.5 Sonnet. Reducing model capacity for a high-risk task is counter-intuitive and increases the likelihood of the agent making a mistake during the execution of a critical operation.
- ✗
Increasing the temperature to allow for more flexibility
Why it's wrong here
Higher temperature increases randomness in the model's output, which is the opposite of what is required for sensitive administrative tasks. For record deletion, you want maximum determinism and predictability, not creative or varied responses that could lead to unexpected behavior or the selection of the wrong records.
- ✓
Human-in-the-Loop (HITL) approval step
Why this is correct
Requiring a human to review and approve the agent's intent before executing a destructive action is the gold standard for agentic security. This ensures that a person validates the context and correctness of the operation, effectively mitigating the risks associated with autonomous decision-making in sensitive production systems.
About these practice questions
This CCAR-P question is part of Courseiva's 262-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam 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 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.