CCARF

CCARF Domain D1

Agentic Architecture & Orchestration Practice Questions

Largest exam domain (~27%). Agentic loops, coordinator–subagent orchestration, Task-tool spawning, hooks for enforcement, session resume/fork, and production workflow handoffs.

50 practice questions · Official blueprint weight 27% · Claude Certified Architect Foundations (CCAR-F)

Paper 1 · Q4A customer support agent needs to escalate a complex billing dispute to a human agent. The AI agent compiles a handoff that includes the conversation summary an… Paper 1 · Q8Production data shows that in 8% of cases, a customer support agent processes refunds without verifying account ownership, occasionally leading to refunds on wr… Paper 1 · Q11A customer support agent handles account deletion requests. The company policy requires explicit human approval before any account is permanently deleted. Curre… Paper 1 · Q12A developer-tool agent writes code and runs shell commands. A system prompt instruction ('Do not run destructive commands or access files outside the project') … Paper 1 · Q15A coordinator agent should spawn specialist subagents for billing disputes and technical issues. Its system prompt describes the delegation workflow correctly, … Paper 1 · Q19An architect is designing a customer support system using the Claude Agent SDK. The system needs a coordinator agent that can delegate to a billing specialist a… Paper 1 · Q32A customer support agent's loop terminates when cumulative token usage exceeds a budget. During a complex billing investigation the budget is exhausted mid-task… Paper 1 · Q35A customer-support agent runs an agentic loop. Each turn it inspects `stop_reason`; when the value is `"tool_use"` it executes the requested tool and receives t… Paper 1 · Q37A coordinator using the Claude Agent SDK delegates a refactoring task to a code reviewer, a test generator, and a documentation writer in parallel, all working … Paper 1 · Q39A customer support agent runs to its `max_iterations=10` cap on every request, even simple ones. Logs show stop_reason returns 'end_turn' after the third iterat… Paper 1 · Q46A customer support agent integrates with three different backend systems via MCP tools. One system returns dates as Unix timestamps, another returns ISO 8601 st… Paper 1 · Q52A developer's customer support agent sometimes terminates prematurely because they check if response.content[0].type == 'text' to determine completion. The agen… Paper 1 · Q53A customer support agent must redact credit card numbers from its responses before they reach the customer. The current system prompt instructs the agent to rep… Paper 1 · Q56A financial services company's customer support agent receives a request: 'I was charged twice for my subscription, and I also need to update my billing address… Paper 2 · Q1A research system receives a customer query: 'Compare the environmental impact and cost- effectiveness of electric vehicles versus hydrogen fuel cell vehicles f… Paper 2 · Q3The team's refactoring pipeline uses a PreToolUse hook to enforce that every new microservice includes a Dockerfile. However, the hook only checks for Dockerfil… Paper 2 · Q8During the monolith-to-microservices refactoring, each extracted service must follow a consistent Java package naming convention (com.company.service.<service-n… Paper 2 · Q13A coordinator spawns a research subagent to investigate market trends. The subagent returns detailed findings. The coordinator then needs to pass these findings… Paper 3 · Q1[Scenario 1: Customer Support Resolution Agent] You are building a customer support resolution agent with the Claude Agent SDK. It handles high-ambiguity reques… Paper 3 · Q2[Scenario 1: Customer Support Resolution Agent] You are building a customer support resolution agent with the Claude Agent SDK. It handles high-ambiguity reques… Paper 3 · Q3[Scenario 1: Customer Support Resolution Agent] You are building a customer support resolution agent with the Claude Agent SDK. It handles high-ambiguity reques… Paper 3 · Q4[Scenario 1: Customer Support Resolution Agent] You are building a customer support resolution agent with the Claude Agent SDK. It handles high-ambiguity reques… Paper 3 · Q5[Scenario 1: Customer Support Resolution Agent] You are building a customer support resolution agent with the Claude Agent SDK. It handles high-ambiguity reques… Paper 3 · Q16[Scenario 2: Code Generation with Claude Code] Your team uses Claude Code for code generation, refactoring, debugging, and documentation, integrated into daily … Paper 3 · Q17[Scenario 2: Code Generation with Claude Code] Your team uses Claude Code for code generation, refactoring, debugging, and documentation, integrated into daily … Paper 3 · Q18[Scenario 2: Code Generation with Claude Code] Your team uses Claude Code for code generation, refactoring, debugging, and documentation, integrated into daily … Paper 3 · Q21[Scenario 3: Multi-Agent Research System] A coordinator agent built on the Claude Agent SDK delegates to specialized subagents — web search, document analysis, … Paper 3 · Q22[Scenario 3: Multi-Agent Research System] A coordinator agent built on the Claude Agent SDK delegates to specialized subagents — web search, document analysis, … Paper 3 · Q23[Scenario 3: Multi-Agent Research System] A coordinator agent built on the Claude Agent SDK delegates to specialized subagents — web search, document analysis, … Paper 3 · Q24[Scenario 3: Multi-Agent Research System] A coordinator agent built on the Claude Agent SDK delegates to specialized subagents — web search, document analysis, … Paper 3 · Q25[Scenario 3: Multi-Agent Research System] A coordinator agent built on the Claude Agent SDK delegates to specialized subagents — web search, document analysis, … Paper 3 · Q38[Scenario 4: Developer Productivity with Claude] You are building developer productivity tooling on the Claude Agent SDK: helping engineers explore unfamiliar c… Paper 3 · Q39[Scenario 4: Developer Productivity with Claude] You are building developer productivity tooling on the Claude Agent SDK: helping engineers explore unfamiliar c… Paper 3 · Q40[Scenario 4: Developer Productivity with Claude] You are building developer productivity tooling on the Claude Agent SDK: helping engineers explore unfamiliar c… Paper 4 · Q6The moderation system handles appeals where users contest a moderation decision. Currently, the same agent that made the original decision re-evaluates the appe… Paper 4 · Q13A multi-agent research system must process a customer's request that involves three sequential stages: data collection, analysis, and report generation. Each st… Paper 4 · Q15In a hub-and-spoke research system, a coordinator manages several search and analysis subagents. Which responsibilities sit with the coordinator rather than wit… Paper 4 · Q23A consulting firm needs an agent to add a comprehensive test suite to a large legacy codebase. The codebase has no existing tests, unclear dependencies between … Paper 4 · Q29An agent occasionally processes international transfers without required compliance checks. The compliance team requires 100% enforcement of anti-money launderi… Paper 4 · Q30The moderation system auto-removes posts classified as policy violations. Production logs reveal that 3% of auto-removed posts were legitimate content (false po… Paper 4 · Q34A developer resumes a Claude Code session after modifying 3 files in a 50-file codebase. The agent gives contradictory advice about the modified files because i… Paper 4 · Q35The content moderation system uses a hub-and-spoke architecture with a coordinator that routes posts to specialist subagents: a text classifier, an image analys… Paper 5 · Q1[Scenario: Multi-agent Research System] A document analysis agent discovers that two credible sources contain directly contradictory statistics for a key metric… Paper 5 · Q2[Scenario: Multi-agent Research System] The web-search and document-analysis agents have completed their tasks and returned results to the coordinator. What is … Paper 5 · Q4[Scenario: Multi-agent Research System] After running the system on “AI impact on creative industries,” you observe that every subagent completes successfully: … Paper 5 · Q8[Scenario: Multi-agent Research System] A colleague proposes that the document analysis agent should send its results directly to the synthesis agent, bypassing… Paper 5 · Q10[Scenario: Multi-agent Research System] While researching a broad topic, you observe that the web-search agent and the document analysis agent investigate the s… Paper 5 · Q13[Scenario: Claude Code for Continuous Integration] Your team uses Claude Code for generating code suggestions, but you notice a pattern: non-obvious issues—perf… Paper 5 · Q35[Scenario: Customer Support Agent] After calling `get_customer` and `lookup_order`, the agent has all available system data but still faces uncertainty. Which s… Paper 5 · Q37[Scenario: Customer Support Agent] Production metrics show your agent averages 4+ API loops per resolution. Analysis reveals Claude often requests `get_customer…

Other domains

D2 Tool Design & MCP Integration · D3 Claude Code Configuration & Workflows · D4 Prompt Engineering & Structured Output · D5 Context Management & Reliability