CCAR-P Advanced Agentic Architecture Practice Test 4 — 15 Questions

Question 1

Which THREE components are essential for building a robust 'human-in-the-loop' (HITL) approval gate within an agentic workflow?

Accepted Answer

A state persistence layer to save the agent context during the pause.. A robust HITL gate requires a state suspension mechanism, a clear communication interface for the human, and an automated resumption process. By pausing the agent's execution, capturing the state, and allowing for human intervention, the system ensures that high-stakes decisions are verified. This is critical for preventing unauthorized or dangerous actions while maintaining the agent's context and momentum throughout the broader automated process.

Answer

A direct feedback loop that forces the model to re-train after every human input.

Answer

A mandatory requirement that humans provide a complete rewrite of the agent's plan.

Question 2

When designing agents that interact with external APIs, which pattern best addresses the challenge of 'unreliable API latency' impacting the agent's reasoning chain?

Accepted Answer

Use an asynchronous 'request-poll' pattern to decouple execution from reasoning.. To manage unpredictable API latency, you must decouple the agent's reasoning from the API execution through an asynchronous task queue. By allowing the agent to request an action and then continue other tasks or enter a wait state, you prevent the reasoning process from timing out or becoming blocked. This is critical for building responsive, reliable agentic systems that can handle real-world network instability and variable service performance.

Answer

Set the agent's request timeout to 60 seconds to ensure it eventually finishes.

Answer

Hardcode a retry count of ten to force the API to respond faster.

Answer

Instruct the model to wait for a response in its thinking process.

Question 3

In a swarm of specialized agents, what is the primary benefit of using a 'Blackboard' pattern for inter-agent communication?

Accepted Answer

It enables decoupled, non-linear collaboration between specialized agents.. The Blackboard pattern allows multiple agents to contribute their expertise to a shared data structure, or blackboard, which serves as a common knowledge base. This promotes decoupling, as agents do not need to know about each other's existence, only how to read from or write to the blackboard. This architecture is essential for complex, multi-faceted problems where multiple specialized agents must collaborate without creating rigid, brittle dependency chains.

Answer

It forces all agents to run sequentially, which improves execution speed.

Answer

It eliminates the need for any form of agent orchestration or management.

Answer

It increases security by isolating agent memory into individual silos.

Question 4

You are building a system that requires strict adherence to a specific output format. Which approach provides the highest reliability in a high-traffic agentic environment?

Accepted Answer

Implement structured schema enforcement at the model interface layer.. Forcing structured output through system-level schema enforcement at the model interface is the only reliable way to guarantee format consistency in production. By utilizing constrained decoding or rigorous post-processing validation, you ensure that the agent consistently produces machine-readable formats like JSON. This architectural pattern is vital because it removes the variability of natural language, allowing downstream systems to process agent outputs without frequent parsing errors or exceptions.

Answer

Include a long prompt instruction asking the model to be very careful with formatting.

Answer

Use a post-generation regex parser to fix up malformed JSON responses.

Answer

Request the model to output the answer in XML because it is easier for agents.

Question 5

Which pattern is most suitable for an agent that must balance the need for speed (latency) versus the need for correctness in a customer support scenario?

Accepted Answer

Use a dual-path architecture with immediate streaming and background verification.. The 'Dual-Path' pattern, where the system initiates an immediate low-latency response while simultaneously running a more comprehensive, high-correctness check in the background, optimizes for both user experience and accuracy. This ensures the user feels acknowledged immediately, while the eventual output remains verified and accurate. This balance is critical in customer support, where perceived responsiveness is as vital as providing correct, verified information for technical or complex queries.

Answer

Always prioritize speed by using the smallest, fastest model available.

Answer

Force the user to wait for verification before showing any response.

Answer

Use a RAG-only architecture to guarantee factual accuracy.

Question 6

You are building a Claude-based agent that must parse unstructured customer emails, extract line-item order data, and then call a fulfillment tool with the extracted values. During testing, the agent occasionally calls the fulfillment tool with empty or garbled line items when an email contains a forwarded message with a different formatting style. Which architectural change most directly reduces this failure?

Accepted Answer

Add a validation step that checks the extracted line items against the tool's JSON schema and returns a structured error to the model before any tool call is allowed.. The failure occurs because extracted line items are not verified before the tool boundary is crossed. A validation gate that compares the extraction against the tool's JSON schema and returns a structured error gives the model a chance to repair its output. This keeps invalid data out of downstream systems and converts a silent corruption into a retryable, observable event.

Answer

Reduce the number of tools available to the agent so it focuses only on fulfillment during extraction.

Answer

Switch the fulfillment tool to accept free-form natural language instead of structured parameters so the model can pass whatever it extracted.

Answer

Increase the max_tokens parameter on the extraction call so the model has more room to emit complete line items.

Question 7

A production agent uses the Claude Messages API with extended thinking enabled to solve multi-constraint scheduling problems. The agent must preserve its reasoning across several tool calls within a single user turn. A developer notices that after the second tool call, the model appears to forget earlier constraints it had already reasoned about. Which change best preserves the reasoning chain across tool calls?

Accepted Answer

Pass the full assistant message, including thinking blocks, back into the conversation history on each subsequent request.. Extended thinking stores reasoning in thinking blocks that are part of the assistant message. When an agent makes multiple tool calls within one turn, each follow-up request must include the prior assistant message with its thinking blocks intact. Omitting them causes the model to lose the constraints it already reasoned through, producing the observed regression after the second tool call.

Answer

Disable extended thinking for subsequent requests once the first tool call has returned, since the constraints are already known.

Answer

Increase the thinking budget_tokens value so the model has more room to reason on each request.

Answer

Summarize the thinking blocks into a single text note and append it as a user message before the next tool call.

Question 8

You are operating a Claude-based support agent that must follow a strict refund policy: refunds over $500 require a manager approval code that is only obtainable through a separate internal API. The agent has access to a `get_manager_code` tool, but in production it occasionally issues refunds above $500 without calling the tool. Which architectural change most reliably prevents this?

Accepted Answer

Add a pre-tool-use hook that inspects the refund amount and blocks any refund tool call above $500 unless a valid manager code is present in the session state.. Hard business constraints must be enforced deterministically outside the model. A pre-tool-use hook inspects the pending refund call and rejects any amount above $500 that lacks a validated manager code, making the violation impossible rather than improbable. Prompting, temperature, and token limits only shift probabilities and cannot guarantee compliance with a policy that carries financial or legal consequences.

Answer

Expand the system prompt to describe the $500 threshold in bold and add two few-shot examples of correct refund handling.

Answer

Increase the max_tokens setting so the agent has more room to reason about whether approval is needed.

Answer

Lower the model temperature to 0 so the agent produces deterministic decisions.

Question 9

You are designing a long-running agent that executes a sequence of irreversible operations, such as issuing refunds and sending customer notifications. The agent runs unattended overnight. Which TWO architectural patterns best ensure that a partial failure does not leave the system in an inconsistent state? (Choose two.)

Accepted Answer

Implement compensating actions so that each irreversible operation has a defined reversal or mitigation step.. Unattended agents performing irreversible actions need both a durable record of intent and a way to reverse or mitigate completed steps. Journaling provides the audit trail and restart logic, while compensating actions provide the semantic undo. Together they allow the orchestrator to detect partial completion and bring the workflow back to a consistent state after a crash or timeout.

Answer

Cache all tool responses in memory so the agent can replay the session without re-calling tools after a restart.

Answer

Raise the max_tokens limit for each planning step so the agent can reason through more contingencies before acting.

Answer

Increase the model's temperature so it explores alternative execution orders and avoids getting stuck on a failing step.

Question 10

A research agent uses Claude with an extended thinking budget to analyze a 200-page regulatory filing. Mid-analysis it must call a `fetch_footnote` tool whose result is essential to the conclusion. The architect wants the tool result to be incorporated without discarding the model's prior reasoning. Which approach best achieves this?

Accepted Answer

Insert the tool result into the conversation as a tool_result block associated with the original tool_use, then continue the same turn so the model resumes from its existing reasoning.. Tool results must be returned as tool_result blocks paired with the originating tool_use so the conversation remains valid and the model can continue from its existing reasoning rather than restart. This preserves the thinking budget and the message sequence. Restarting, summarizing through another model, or deferring to retrieval all either discard reasoning or break the API contract, so they fail this scenario.

Answer

Store the footnote in an external vector store and instruct the agent to retrieve it again later if needed.

Answer

Append the tool result as a new user message and restart the analysis from the beginning of the filing.

Answer

Summarize the footnote with a separate cheap model and paste the summary into the system prompt before the next call.

Question 11

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?

Accepted Answer

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

Fine-tune a separate model on historical tickets and deploy it alongside the Claude agent for classification.

Answer

Give the agent access to all internal tools and let it infer the category from whichever tool it chooses to call.

Answer

Build a multi-agent pipeline where a classifier agent debates a router agent until they agree on a category.

Question 12

You are architecting a Claude agent that orchestrates a long-running approval workflow spanning hours or days, where human reviewers may intervene between steps. Which TWO mechanisms are necessary to keep the workflow correct across these interruptions? (Choose two.)

Accepted Answer

Persist the full message history and tool state to durable storage after each turn so the workflow can resume from the exact checkpoint.. Long-running workflows with human-in-the-loop interruptions require durable checkpointing to resume exact state and idempotency keys to make retries safe. Together they ensure that an interruption or crash neither loses the reasoning chain nor duplicates side effects such as approvals or notifications. Stateless summarization, token limits, and disabling tools all fail to provide either guarantee.

Answer

Set a very short max_tokens on every turn to reduce the chance of context drift across days.

Answer

Disable tool use entirely during human review windows and rely on the reviewer to re-enter all prior context manually.

Answer

Use a stateless design where each invocation rebuilds context by summarizing all prior decisions into the system prompt.

Question 13

You are building a customer-support agent with Claude that must call a lookup_order tool, then a refund_order tool that depends on the order's status. During testing, the agent sometimes calls refund_order before the lookup_order result returns. Which change to your orchestration loop best enforces the required ordering?

Accepted Answer

Expose only lookup_order until its result is returned, then add refund_order to the tools array for the subsequent request.. Deterministic ordering is achieved by controlling what the model can do at each step, not by prompting harder. Withholding the dependent tool until the prerequisite result is in the conversation removes the failure mode entirely, because the model cannot select a tool absent from the current tools array. This makes the dependency a property of the orchestration state machine rather than of model compliance.

Answer

Increase the max_tokens value so the model has more room to reason about the correct order before calling tools.

Answer

Include both tools in the same request's tools array and let the model choose the call order.

Answer

Set tool_choice to force a tool call on every turn so the agent never stalls between lookup and refund.

Question 14

An agent uses a retrieval tool that returns the top 20 chunks for any query. In production, the agent frequently cites irrelevant chunks and sometimes misses the correct answer even when it is present in the corpus. The corpus contains documents with overlapping terminology. Which architectural change most improves answer grounding without increasing the number of retrieved chunks?

Accepted Answer

Add a re-ranking stage that scores the retrieved chunks against the query and passes only the highest-scoring subset to the model.. The retrieval step is high-recall but low-precision, which is why irrelevant chunks are cited and correct ones are missed amid overlapping terminology. Adding a re-ranking stage that scores candidates against the query and passes only the top subset improves precision without retrieving more chunks, directly improving grounding and citation quality.

Answer

Embed the entire corpus into the system prompt so the model always has full context available.

Answer

Increase the retrieval count to 50 so the correct chunk is more likely to be included somewhere in the context.

Answer

Lower the model's temperature to zero so it sticks more closely to the retrieved text.

Question 15

An orchestration agent runs a nightly workflow that fans out to 12 subagents, each calling a partner REST API. Partner calls intermittently return HTTP 429 with a Retry-After header. The orchestrator currently retries immediately in a tight loop, causing cascading 429s and duplicate side effects on the partner systems. Which architectural change best addresses both the throttling and the duplicate side effects?

Accepted Answer

Introduce a shared token-bucket rate limiter in front of the partner calls plus idempotency keys on every mutating request, with subagents respecting Retry-After.. Bursty fan-out against a rate-limited partner needs two coordinated controls: a shared limiter so the aggregate request rate stays under the partner's ceiling, and idempotency keys so a retried mutating call cannot create a second side effect. Honoring Retry-After aligns retry timing with the partner's guidance. Token or context changes operate on model text, not on outbound traffic, and serializing the workload removes parallelism without addressing throttling or duplication.

Answer

Reduce the fan-out to a single subagent that processes all 12 partners sequentially with no retry logic at all.

Answer

Switch every subagent to a larger context window so each can hold the full history of prior 429 responses and learn to avoid them.

Answer

Increase the orchestrator's max_tokens so the model can reason longer about each 429 before deciding to retry.