CCAR-P Advanced Agentic Architecture Practice Question
A Claude agent maintains a long conversation with a user over weeks. The architect notices that the agent gradually forgets early constraints the user stated, even though the conversation is well within the model's context window. The team wants to fix this without re-summarizing the entire history on every turn. Which approach is most appropriate?
⚠ Common exam trap
The trap here is assuming that a larger context window solves forgetting, when the scenario already excludes that and the real issue is salience of persistent constraints.
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
✓
Move persistent user constraints into a structured memory store that is retrieved and injected as a system-level block on each turn.
Persistent constraints belong in a durable memory store that is retrieved and injected as a system-level block each turn, giving them stable salience without repeatedly summarizing the whole history. Larger context windows, repeated reminders, and temperature changes do not fix attention dilution over long conversations, and two of them are actively counterproductive. Structured memory is the right architectural layer for durable user preferences.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Raise the temperature slightly so the model explores more of the conversation when generating.
Why it's wrong here
Temperature controls randomness in token selection, not attention to earlier content. Raising it would make responses less consistent and more prone to ignoring constraints, not more faithful. This misdiagnoses the problem as a sampling issue when it is really a context-salience issue.
- ✗
Increase the model's context window by switching to a larger variant so all history fits with room to spare.
Why it's wrong here
The scenario states the conversation is already within the context window, so a larger window does not address the root cause, which is attention dilution over long histories. Bigger windows also cost more and still suffer recency bias. This treats a symptom the scenario explicitly rules out.
- ✓
Move persistent user constraints into a structured memory store that is retrieved and injected as a system-level block on each turn.
Why this is correct
Extracting durable constraints into a structured memory store and re-injecting them each turn keeps them salient regardless of how the conversation grows, without re-summarizing everything. It directly addresses gradual forgetting by giving constraints a stable, high-priority position. This is the scalable pattern for long-lived agents with persistent user preferences.
- ✗
Insert the early constraints again at the very end of every user message as a reminder.
Why it's wrong here
Repeating constraints in every user turn is brittle, bloats the prompt, and can confuse the model about who said what. It also fails to distinguish durable preferences from transient instructions. A structured store is cleaner and more reliable than ad hoc repetition, which is why this approach is inferior.
About these practice questions
Courseiva writes every CCAR-P question from scratch — 262 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 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.