CCAR-P Advanced Agentic Architecture Practice Question
A support agent built on Claude must decide whether a customer request is a billing issue, a technical issue, or a general inquiry before routing it. The categories are fixed and mutually exclusive, and the routing decision must be fast and cheap. Which approach is most appropriate?
⚠ Common exam trap
The trap here is equating higher architectural complexity with higher accuracy, when a constrained single call already guarantees a valid label for a fixed taxonomy.
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
✓
Use a single Claude call with a constrained output format such as a tool definition or structured JSON that returns exactly one category label.
Fixed, mutually exclusive categories with strict latency and cost goals are best handled by a single Claude call that constrains output to one of the allowed labels. Using a tool definition or JSON schema with an enum removes ambiguity and avoids the overhead of multi-agent orchestration or a separate fine-tuned model.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Use a single Claude call with a constrained output format such as a tool definition or structured JSON that returns exactly one category label.
Why this is correct
A fixed, mutually exclusive classification with speed and cost constraints is a natural fit for a single constrained call. Defining a tool or JSON schema with an enum of the three categories forces the model to emit one valid label. This avoids multi-step orchestration overhead and keeps latency and token cost low, which matches the stated requirements exactly.
- ✗
Fine-tune a separate model on historical tickets and deploy it alongside the Claude agent for classification.
Why it's wrong here
Fine-tuning introduces training data preparation, deployment, and maintenance overhead that is disproportionate for a three-way fixed classification. A constrained prompt or tool definition achieves the same result without a second model lifecycle. The scenario emphasizes speed and cost, and adding a fine-tuned model increases both operational complexity and infrastructure cost. It is not the most appropriate first solution.
- ✗
Give the agent access to all internal tools and let it infer the category from whichever tool it chooses to call.
Why it's wrong here
Inferring category from tool selection conflates two decisions and makes the routing label an implicit side effect rather than an explicit output. It also exposes the agent to unnecessary tools, increasing the chance of an unrelated call. The requirement is an explicit, reliable category label, which this approach does not guarantee. It is indirect and harder to audit.
- ✗
Build a multi-agent pipeline where a classifier agent debates a router agent until they agree on a category.
Why it's wrong here
A debate between two agents adds multiple model calls, coordination logic, and latency for a task that a single constrained call can handle. Mutual agreement is not required for a fixed taxonomy, and the extra complexity increases cost without improving accuracy meaningfully. This over-engineers a straightforward classification and contradicts the requirement for fast, cheap routing.
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.