quizzMonkey
Start free
CCAR-FAnthropicExam Guide v1.0 (effective July 2026)

Claude Certified Architect – Foundations

60 scenario questions across 5 domains in 120 minutes, scored 100 to 1,000 with a 720 pass mark: the same format as the official exam. Every option is explained.

60 questions120 minutesPass 720 / 1,0005 domains192 practice questionsOfficial exam page

Syllabus

5 domains · 30 skills, weighted like the official blueprint.

1Agentic Architecture & Orchestration27% of the exam · 7 skills
  • 1.1 Design and implement agentic loops for autonomous task executionNot started
    3.9% · 6 questions
    Notes
  • 1.2 Orchestrate multi-agent systems with coordinator-subagent patternsNot started
    3.9% · 7 questions
    Notes
  • 1.3 Configure subagent invocation, context passing, and spawningNot started
    3.9% · 6 questions
    Notes
  • 1.4 Implement multi-step workflows with enforcement and handoff patternsNot started
    3.9% · 7 questions
    Notes
  • 1.5 Apply Agent SDK hooks for tool call interception and data normalizationNot started
    3.8% · 6 questions
    Notes
  • 1.6 Design task decomposition strategies for complex workflowsNot started
    3.8% · 6 questions
    Notes
  • 1.7 Manage session state, resumption, and forkingNot started
    3.8% · 6 questions
    Notes
2Tool Design & MCP Integration18% of the exam · 5 skills
  • 2.1 Design effective tool interfaces with clear descriptions and boundariesNot started
    3.6% · 7 questions
    Notes
  • 2.2 Implement structured error responses for MCP toolsNot started
    3.6% · 6 questions
    Notes
  • 2.3 Distribute tools appropriately across agents and configure tool choiceNot started
    3.6% · 7 questions
    Notes
  • 2.4 Integrate MCP servers into Claude Code and agent workflowsNot started
    3.6% · 6 questions
    Notes
  • 2.5 Select and apply built-in tools (Read, Write, Edit, Bash, Grep, Glob) effectivelyNot started
    3.6% · 6 questions
    Notes
3Claude Code Configuration & Workflows20% of the exam · 6 skills
  • 3.1 Configure CLAUDE.md files with appropriate hierarchy, scoping, and modular organizationNot started
    3.3% · 6 questions
    Notes
  • 3.2 Create and configure custom slash commands and skillsNot started
    3.3% · 7 questions
    Notes
  • 3.3 Apply path-specific rules for conditional convention loadingNot started
    3.3% · 7 questions
    Notes
  • 3.4 Determine when to use plan mode vs direct executionNot started
    3.3% · 7 questions
    Notes
  • 3.5 Apply iterative refinement techniques for progressive improvementNot started
    3.4% · 6 questions
    Notes
  • 3.6 Integrate Claude Code into CI/CD pipelinesNot started
    3.4% · 7 questions
    Notes
4Prompt Engineering & Structured Output20% of the exam · 6 skills
  • 4.1 Design prompts with explicit criteria to improve precision and reduce false positivesNot started
    3.3% · 6 questions
    Notes
  • 4.2 Apply few-shot prompting to improve output consistency and qualityNot started
    3.3% · 6 questions
    Notes
  • 4.3 Enforce structured output using tool use and JSON schemasNot started
    3.3% · 6 questions
    Notes
  • 4.4 Implement validation, retry, and feedback loops for extraction qualityNot started
    3.3% · 6 questions
    Notes
  • 4.5 Design efficient batch processing strategiesNot started
    3.4% · 7 questions
    Notes
  • 4.6 Design multi-instance and multi-pass review architecturesNot started
    3.4% · 7 questions
    Notes
5Context Management & Reliability15% of the exam · 6 skills
  • 5.1 Manage conversation context to preserve critical information across long interactionsNot started
    2.5% · 6 questions
    Notes
  • 5.2 Design effective escalation and ambiguity resolution patternsNot started
    2.5% · 7 questions
    Notes
  • 5.3 Implement error propagation strategies across multi-agent systemsNot started
    2.5% · 7 questions
    Notes
  • 5.4 Manage context effectively in large codebase explorationNot started
    2.5% · 6 questions
    Notes
  • 5.5 Design human review workflows and confidence calibrationNot started
    2.5% · 6 questions
    Notes
  • 5.6 Preserve information provenance and handle uncertainty in multi-source synthesisNot started
    2.5% · 6 questions
    Notes

Sample questions with answers

10 questions from the CCAR-F bank, with an explanation for every option. No sign-up needed.

See sample questions

Questions about this exam

How many questions are in the CCAR-F practice exam, and how long is it?

60 questions in 120 minutes. Each mock draws questions from every domain in proportion to its weight, so the mix matches the blueprint.

What score do you need to pass CCAR-F?

Results are reported from 100 to 1,000, and 720 is the passing score used in these mocks. Harder questions carry more weight.

What does the CCAR-F exam cover?

5 domains: Agentic Architecture & Orchestration (27%), Tool Design & MCP Integration (18%), Claude Code Configuration & Workflows (20%), Prompt Engineering & Structured Output (20%) and Context Management & Reliability (15%).

How can I practise for it?

Start with the free 15-question diagnostic to find your weak spots, drill any skill with an explanation after every answer, retry your mistakes with spaced review, then sit timed full mocks.

How is my readiness score worked out?

It weighs your recent answers on each skill by difficulty, combines the skills by their exam weight into a predicted score, and reports your chance of reaching the pass mark. How scoring works

Is this the official CCAR-F exam?

No. quizzMonkey is an independent practice platform and isn't affiliated with Anthropic. Practise here, then book the real exam with the vendor. Readiness certificates are not official credentials. Official exam page