The official exam

Questions

60

Time

120 minutes

Fee

US$125

Passing score

720 on a 100-1000 scale

Valid for

12 months

Format

Scenario-based multiple-choice and multiple-response; four scenarios drawn from six

Delivery

Proctored online or at a test center

As published by Anthropic in exam guide 1.0 (July 2026); checked 9 August 2026. Read the official exam guide

Who it's for

Developers and architects who design production Claude systems and can reason through ambiguous, scenario-based architecture decisions.

• Six or more months building with Claude
• Hands-on work with the Claude API and tool use
• Working knowledge of Claude Code, the Claude Agent SDK, and MCP
• Experience choosing between architectural alternatives under constraints

What the exam covers

• Agentic Architecture and Orchestration27%
• Tool Design and MCP Integration18%
• Claude Code Configuration and Workflows20%
• Prompt Engineering and Structured Output20%
• Context Management and Reliability15%

Practice exams

Original questions in the official format, timed like the real thing, with a breakdown by exam domain and an explanation for every answer. Start with the diagnostic.

• Architect Foundations Diagnostic15 questions, 30 minutesFree
• Architect Foundations Scenario Mock60 questions, 120 minutes

Lessons

1. 1Study the Decisions, Not the VocabularyA certification blueprint is a map of decisions a competent practitioner can defend. Treat it as a list of terms and you will study the least useful part of the exam.LearnPython1.5 h
2. 2Choose the Smallest Surface That Can Carry the WorkProduct selection is architecture at knowledge-work scale. The wrong surface can make correct output stale, unreviewable, or needlessly expensive.LearnPython1.5 h
3. 3Spend Capability Where Failure Is ExpensiveModel selection is not a ranking exercise. It is an allocation problem across quality, latency, context, and cost.LearnPython1.5 h
4. 4Turn a Request Into a Testable ContractA strong prompt does not merely describe what to write. It makes success observable before generation begins.LearnPython1.5 h
5. 5Put Each Fact in the Right Kind of ContextContext is temporary attention. Knowledge is maintained evidence. Memory is continuity. Caching is reuse. Mixing them creates confident stale answers.LearnPython2 h
6. 6Validate the Claim, Not the ConfidenceFluency is presentation quality. Validation is evidence that the output can safely do its job.LearnPython2 h
7. 7The Messages API Is a State MachineThe API does not remember your conversation. Your application does, and one misplaced content block can break the entire loop.BuildPython2 h
8. 8Structured Output Is an Untrusted ContractValid JSON is not a valid business decision. Parse the bytes, validate the shape, verify the meaning, then permit the action.BuildPython1.5 h
9. 9A Tool Loop Is Controlled DelegationClaude may propose an action. Your application validates the request, grants the capability, observes the result, and decides whether the loop continues.BuildPython2 h
10. 10MCP Separates Capability From HostBuild a narrow, stateless MCP server whose contract can be discovered, cached, invoked, and scaled without hidden connection state.BuildPython2 h
11. 11The Agent SDK Is a Harness, Not PermissionAn agent becomes dependable when the loop, tools, context, hooks, and termination policy are explicit enough to inspect and constrain.LearnPython2.5 h
12. 12Security Lives Outside the PromptThe model can recommend a safe action. Only deterministic controls can make an unsafe action impossible.BuildPython2 h
13. 13Evals Turn Agent Behavior Into Engineering EvidenceA trace tells you what happened. An eval tells you whether it was acceptable. A regression gate keeps the next change from quietly making it worse.BuildPython2 h
14. 14Claude Code Scales Through Shared ConstraintsA team does not need one giant prompt. It needs a small project contract, reusable procedures, deterministic checks, and versioned configuration.LearnPython3 h
15. 15Multi-Agent Orchestration and DelegationDelegate a bounded question, not your entire uncertainty.ReferencePython2.5 h
16. 16Agent SDK Sessions, Subagents, and ContextResume state when continuity helps. Fork context when inherited assumptions become risk.ReferencePython2 h
17. 17Tool Contracts, Errors, and Progressive DiscoveryThe model chooses from the interface you describe. Ambiguous tools create ambiguous behavior.ReferencePython2 h
18. 18Claude Code Memory, Rules, Skills, and CIPut stable guidance where its scope is true, and executable constraints where failure is unacceptable.ReferencePython3.5 h
19. 19Reliable Extraction, Batch, and Independent ReviewersValid JSON proves the shape survived. It does not prove the facts did.ReferencePython2.5 h
20. 20Make Large Context ObservableA large context window can hold more evidence. It cannot tell you which evidence was noticed, current, authoritative, or safe to act on.ReferencePython2.5 h
21. 21Defend One Architecture Across Six ContextsArchitecture is the set of boundaries that still hold when the scenario changes, a tool fails, and the evidence is incomplete.BuildPython6 h

Study plans

Go deeper in the Academy

• Orchestration Patterns (Phase 14): Compare coordinator, delegation, and parallel patterns
• Durable Execution (Phase 15): Handle resume, failure, and long-running state
• MCP Tool Poisoning (Phase 13): Treat tool descriptions and results as an attack surface
• Reviewer Agent (Phase 14): Build independent evidence-based review
• Multi-Session Handoff (Phase 14): Preserve structured state without stale context

Katalock's Claude certification prep is independent study material based on Anthropic's public exam guides. It is not affiliated with, endorsed by or authorised by Anthropic, does not contain live exam questions, and cannot guarantee a pass. Practice scores are raw percentages. They are not Anthropic's 100 to 1,000 scaled score and can't predict an official result.