Anthropic · CCAR-P
Claude Certified Architect Professional (CCAR-P) Practice Tests
Validates advanced expertise in designing and architecting production-grade AI solutions using Claude, targeting solution architects, AI/ML engineers, and technical leads with 3+ years of systems-architecture experience. Covers solution design, integration patterns (MCP, API, agent-to-agent), evaluation, governance, safety, and operational enablement.
Exam Details
Practice Questions
306
≈ 4 practice exams
Duration
120 minutes
Passing Score
720/1000
Difficulty
ProfessionalLast Updated
Sep 2026
Topics Covered
Solution Design & ArchitectureClaude Models, Prompting & Context EngineeringIntegration Mechanisms (MCP, API/CLI, Agent-to-Agent)Evaluation, Testing & OptimisationGovernance, Safety & Risk ManagementStakeholder CommunicationDeveloper Productivity & Operational Enablement
Claude Certified Architect Professional (CCAR-P) Practice Exam Preparation
Use this CCAR-P practice exam to prepare for Claude Certified Architect Professional (CCAR-P) with realistic questions, detailed explanations, and focused study modes. The practice bank includes 306 questions for Anthropic CCAR-P, so you can review the exam steadily instead of relying on one long cram session.
As you practice, pay extra attention to recurring topics such as Solution Design & Architecture, Claude Models, Prompting & Context Engineering, Integration Mechanisms (MCP, API/CLI, Agent-to-Agent), Evaluation, Testing & Optimisation, and Governance, Safety & Risk Management. Start with short sessions to identify weak areas, then move into timed quizzes once your accuracy is consistent.
The explanations are written for candidates preparing for Anthropic certification questions and reviewing how the concepts apply in practice. Use the free preview first, then unlock the full question bank when you are ready to build a complete study routine.
Sample Questions
5 sample questions with answers and explanations. The full bank has 306 questions, enough for 4 full-length practice exams.
Preview — answers shown1. A team is considering upgrading to a larger model to improve performance. Which statement about data exposure is correct? (Select one!)
AIt does not change data exposure or access scope
BIt reduces data exposure
CIt increases access scope
DIt changes the authorization model
Explanation
Model size has nothing to do with authorization; using a bigger or better model does not change data exposure or access scope.
2. You are configuring file access for Claude Cowork. Which set of file-mount modes does it offer? (Select one!)
ARead-only, read-write, and no-access
BRead-only, read-write, and execute-only
CRead-only, read-write, and read-write-no-delete
DRead-only, write-only, and read-write
Explanation
Claude Cowork offers different file-mount modes: read-only, read-write, and read-write-no-delete. These modes control the level of access Claude has to the mounted workspace. The other options either omit a mode or include a mode that is not offered.
3. What is required for effective sandboxing of an agent? (Select one!)
AFilesystem isolation alone is sufficient for sandboxing
BBoth filesystem and network isolation are required
COnly outbound connections need to be blocked
DNetwork isolation is optional if the agent is trusted
Explanation
Effective sandboxing requires both filesystem and network isolation. Without network isolation, a compromised agent could exfiltrate sensitive files like SSH keys. Filesystem isolation alone, relying on the agent's honesty, or only blocking outbound connections would not prevent exfiltration.
4. What is the isolation scope of Claude Cowork's virtual machine? (Select one!)
AThe VM runs as a container on the host
BThe entire host filesystem is mounted into the VM
CCredentials are stored inside the guest machine
DOnly the selected workspace and `.claude` folder are mounted
Explanation
Claude Cowork runs inside a full virtual machine using the platform's vendor hypervisor. Only the user's selected workspace and `.claude` folder are mounted; nothing else on the host is visible, and credentials stay in the host's keychain. Mounting the entire host filesystem, using a container, or storing credentials in the guest would violate the isolation model.
5. What data does the Claude Code auto mode transcript classifier use to make decisions, and why? (Select one!)
AUser messages, Claude's messages, and tool inputs
BClaude's messages and tool outputs only
CUser messages and the agent's tool calls only
DUser messages, Claude's messages, and tool outputs
Explanation
The Claude Code auto mode transcript classifier is intentionally reasoning-blind: it sees only user messages and the agent's tool calls, with Claude's own messages and tool outputs stripped out. This design ensures the classifier evaluates whether an action aligns with the user's intent based on what the user asked for and what the agent is doing, not on Claude's intermediate reasoning or the results of prior actions. Therefore, options that include Claude's messages (A, C) or tool outputs (A, D) are incorrect.
$17.99
One-time access to this exam
306 questions (4 practice exams' worth)
Unlimited timed exam simulations
Yours forever — no subscription
Detailed explanations
Free preview stays available