Anthropic · CCAR-F
Claude Certified Architect: Foundations (CCAR-F) — Practice Course
The full CCAR-F blueprint: six production scenarios, 240 questions, and written study notes for all thirty task statements.
240 practice questions · Foundational
What you’ll learn
- Sit a full 60-question, 120-minute exam simulation under real conditions and find out whether you are ready at the 72% cut
- Decide when a coordinator-subagent split earns its cost, and when a single agentic loop is the better design
- Place a guarantee where it will actually hold — in the tool layer — instead of trusting a system prompt to enforce it
- Scope CLAUDE.md, slash commands, skills and plan mode so a whole team gets consistent results from Claude Code
- Enforce structured output with JSON schemas and tool use, then validate, retry and batch it at production scale
- Manage context, escalation and error propagation so long multi-agent sessions stay reliable and traceable
- Pinpoint weak areas by domain and task statement, so your revision targets the blueprint rather than guesswork
About this exam
This course is built directly from Anthropic's published Claude Certified Architect — Foundations exam guide. It covers all five domains, all thirty task statements, and all six production scenarios the exam draws from: a customer support resolution agent, code generation with Claude Code, a multi-agent research system, developer productivity tooling, Claude Code in CI/CD, and structured data extraction.
The real exam presents four of those six scenarios, picked at random, and every question is multiple choice with one correct answer and three distractors. This course mirrors that. Sit any single scenario as a focused forty-question test, or take the full exam simulation, which picks four scenarios at random and draws forty questions weighted to the published domain percentages.
Start with the course material. Each of the thirty task statements gets its own section covering what the blueprint expects you to know, what it expects you to be able to do, and the anti-patterns that show up as distractors. Your results link straight back to the sections you got wrong, so a score becomes a study plan.
Note on scoring: the real exam reports a scaled score from 100 to 1,000 with 720 to pass. This test scores by percentage, so the pass mark is shown as 72%.
Course material
Written study notes for every task statement on the blueprint — what to know, what to be able to do, and the anti-patterns that show up as distractors.
Course content
5 sections · 30 topics
Domain 1 — Agentic Architecture & Orchestration (27%)7 topics
- 1.1 Design and implement agentic loops for autonomous task execution
- 1.2 Orchestrate multi-agent systems with coordinator-subagent patterns
- 1.3 Configure subagent invocation, context passing, and spawning
- 1.4 Implement multi-step workflows with enforcement and handoff patterns
- 1.5 Apply Agent SDK hooks for tool call interception and data normalization
- 1.6 Design task decomposition strategies for complex workflows
- 1.7 Manage session state, resumption, and forking
Domain 2 — Tool Design & MCP Integration (18%)5 topics
- 2.1 Design effective tool interfaces with clear descriptions and boundaries
- 2.2 Implement structured error responses for MCP tools
- 2.3 Distribute tools appropriately across agents and configure tool choice
- 2.4 Integrate MCP servers into Claude Code and agent workflows
- 2.5 Select and apply built-in tools (Read, Write, Edit, Bash, Grep, Glob) effectively
Domain 3 — Claude Code Configuration & Workflows (20%)6 topics
- 3.1 Configure CLAUDE.md files with appropriate hierarchy, scoping, and modular organization
- 3.2 Create and configure custom slash commands and skills
- 3.3 Apply path-specific rules for conditional convention loading
- 3.4 Determine when to use plan mode vs direct execution
- 3.5 Apply iterative refinement techniques for progressive improvement
- 3.6 Integrate Claude Code into CI/CD pipelines
Domain 4 — Prompt Engineering & Structured Output (20%)6 topics
- 4.1 Design prompts with explicit criteria to improve precision and reduce false positives
- 4.2 Apply few-shot prompting to improve output consistency and quality
- 4.3 Enforce structured output using tool use and JSON schemas
- 4.4 Implement validation, retry, and feedback loops for extraction quality
- 4.5 Design efficient batch processing strategies
- 4.6 Design multi-instance and multi-pass review architectures
Domain 5 — Context Management & Reliability (15%)6 topics
- 5.1 Manage conversation context to preserve critical information across long interactions
- 5.2 Design effective escalation and ambiguity resolution patterns
- 5.3 Implement error propagation strategies across multi-agent systems
- 5.4 Manage context effectively in large codebase exploration
- 5.5 Design human review workflows and confidence calibration
- 5.6 Preserve information provenance and handle uncertainty in multi-source synthesis
Domain weighting
The practice test draws questions in these proportions, and scores you against them at the end.