Claude Certified Architect exam questions, with answers explained
Claude Certified Architect exam questions are scenario-based: a situation, four options, one best answer. Here are samples for each of the five domains, each with the answer and the reasoning, then a quick-check you can score.
We post a fresh 4-question sample each week - the same set all week (it stays put on reload), with new questions next week so you can track your progress.
60 questions, 120 minutes, 720 of 1000 to pass.
3-attempt lifetime cap. Five weighted domains. The quick check above samples from the four highest-weighted (skipping context + reliability for length).
What the questions look like
Each item gives you a production situation and asks for the best architectural choice. The answer is the option that follows Claude's documented control signals and constraints, not the option that merely sounds plausible.
Sample questions by domain
D1 · Your agentic loop keeps running after Claude has clearly finished its task. Which control was likely missed?
The harness was checking content shape (text presence) instead of stop_reason. The most-tested distractor is "add a max_iterations cap". The right answer is "branch the loop on stop_reason 'end_turn'", structural field, never natural language.
Review Agentic LoopsD2 · Your agent calls the wrong tool 30% of the time across 8 similar tools. What's the first fix?
Audit tool descriptions. Add the four-part anatomy: what it does, when to use, edge cases, ordering boundaries. Vague descriptions cause misrouting; bigger models don't fix it.
Review Tool CallingD3 · A teammate's PRs use a different code style than yours. They have rules in ~/.claude/CLAUDE.md but not in the repo. Fix?
Move team rules to .claude/CLAUDE.md (project root, version-controlled). User-level rules are personal; only project rules are shared via git. Team standards belong in the repo.
Review CLAUDE.md HierarchyD4 · System prompt says "be conservative with refunds." Production shows 5% policy violations. Why?
Natural-language guidance is advice the model ignores under pressure. Encode the policy in tool code (if amount > 500: deny). System prompt = macro guidance; tools = enforcement.
Review System Prompts & InstructionsD5 · Context window resets after each turn?
No. Cumulative across all turns. Messages accumulate; budget shrinks. Once you hit 200K total, new requests fail with context_length_exceeded.
Review Context Window ManagementMCP (Model Context Protocol) questions
MCP exam questions here mean Model Context Protocol in CCAR-F domain 2. These samples cover tool design, MCP servers, error contracts, and context cost.
1. Your agent calls the wrong tool 30% of the time across 8 similar tools. What's the first fix?
Audit tool descriptions. Add the four-part anatomy: what it does, when to use, edge cases, ordering boundaries. Vague descriptions cause misrouting; bigger models don't fix it.
Read Tool Calling2. You add 25 tools and selection accuracy drops sharply. Why?
Beyond ~18 tools, attention dilutes and Claude can't reliably select. Split into specialized subagents, each with 4-5 tools max. Or merge overlapping tools.
Read Tool Calling3. A tool's input_schema is {data: any} and Claude returns null. What happened?
The schema gave no constraint, so Claude has no deterministic target. Model guessed null. Specify exact fields with types and examples; constraints reduce hallucination.
Read Tool Calling4. Your tool throws an exception. The agent retries the same call with the same input. Why?
Your harness swallowed the exception and returned an empty tool_result. Claude saw no error and re-requested. Return {error: "reason", hint: "how to fix"} so Claude can recover.
Read Tool Calling5. Two tools have similar names: fetch_user and get_user. The agent alternates between them. Fix?
Disambiguate in descriptions: "fetch_user: use only when you have an email. get_user: use when you have a numeric ID." Or merge them into one tool with an enum input.
Read Tool Calling6. You force tool_choice: any to guarantee structured output. The agent still returns empty results. Why?
any forces a tool call but the model still picks which one, and might pick wrong if descriptions overlap. Either force a specific tool or fix description disambiguation.
Read Tool Calling7. Your tool returns 50KB of JSON per call. The loop hits max_tokens after 3 iterations. Architectural fix?
Tools should return what's needed, not everything. Add a fields parameter to the tool's input schema so Claude can request specific fields. Or use a resource (read-only catalog) for catalog data.
Read Tool Calling8. A tool needs the customer ID but Claude calls it without one. The harness errors out. What should the schema enforce?
Mark customer_id as required in input_schema. Claude won't call without it. Combine with a clear description: "call only after verify_customer returns a customer_id."
Read Tool CallingRead the Model Context Protocol concept before another attempt.
Questions for the other certifications
Use the CCAR-P quick-check, CCDV-F quick-check, or CCAO-F quick-check for a certification-specific sample.
Why these are original practice questions
These questions are written for learning. They test the public exam blueprint and documented Claude behavior, then explain the reasoning so you can apply the same principle to a new scenario.