Skip to main content
CCAR-F Exam Prep
  • Dashboard
  • Start here
  • Domains
  • Practice
  • Exams
  • Study
  • Labs
  • Resources

App settings

Preferences

Theme

Display and performance

Visual effects

Reduced effects stop ambient drift and shorten transitions.

Reading size

Used only to plan your study list.

Focused Review scope

Number of tasks in a combined review session.

These settings are saved only in this browser. We don't create an account.

Loading…

AboutGlossaryResourcesPrivacy

Authored by Daniel-John Diala and William Judd.

Independent and unofficial. Not affiliated with or endorsed by Anthropic. Practice questions and labs are original, not live-exam content.

The blueprint

Five domains and 30 task statements at their published exam weights.

Jump to instructional diagrams (11) ↓

Domain 1 — Agentic Architecture & Orchestration

27% of the exam · 7 task statements · 7 lessons available

Agentic loops, coordinator-subagent orchestration, task decomposition, hooks, and session state.

  1. 1.1Design and implement agentic loops for autonomous task execution6 questions
  2. 1.2Orchestrate multi-agent systems with coordinator-subagent patterns6 questions
  3. 1.3Configure subagent invocation, context passing, and spawning6 questions
  4. 1.4Implement multi-step workflows with enforcement and handoff6 questions
  5. 1.5Apply Agent SDK hooks for tool call interception and data flow6 questions
  6. 1.6Design task decomposition strategies for complex workflows6 questions
  7. 1.7Manage session state, resumption, and forking6 questions

Practice Domain 1

Domain 2 — Tool Design & MCP Integration

18% of the exam · 5 task statements · 5 lessons available

Tool interfaces, structured errors, tool distribution and choice, MCP servers, and built-in tools.

  1. 2.1Design effective tool interfaces with clear descriptions and boundaries6 questions
  2. 2.2Implement structured error responses for MCP tools6 questions
  3. 2.3Distribute tools appropriately across agents and configure tool choice6 questions
  4. 2.4Integrate MCP servers into Claude Code and agent workflows6 questions
  5. 2.5Select and apply built-in tools (Read, Write, Edit, Bash, Grep, Glob)6 questions

Practice Domain 2

Domain 3 — Claude Code Configuration & Workflows

20% of the exam · 6 task statements · 6 lessons available

CLAUDE.md hierarchy, skills and slash commands, path-scoped rules, plan mode, and CI/CD.

  1. 3.1Configure CLAUDE.md files with appropriate hierarchy, scoping, and organization6 questions
  2. 3.2Create and configure custom slash commands and skills6 questions
  3. 3.3Apply path-specific rules for conditional convention loading6 questions
  4. 3.4Determine when to use plan mode vs direct execution6 questions
  5. 3.5Apply iterative refinement techniques for progressive improvement6 questions
  6. 3.6Integrate Claude Code into CI/CD pipelines6 questions

Practice Domain 3

Domain 4 — Prompt Engineering & Structured Output

20% of the exam · 6 task statements · 6 lessons available

Precision criteria, few-shot prompting, JSON schemas via tool use, validation loops, and batching.

  1. 4.1Design prompts with explicit criteria to improve precision and reduce false positives6 questions
  2. 4.2Apply few-shot prompting to improve output consistency and quality6 questions
  3. 4.3Enforce structured output using tool use and JSON schemas6 questions
  4. 4.4Implement validation, retry, and feedback loops for extraction6 questions
  5. 4.5Design efficient batch processing strategies6 questions
  6. 4.6Design multi-instance and multi-pass review architectures6 questions

Practice Domain 4

Domain 5 — Context Management & Reliability

15% of the exam · 6 task statements · 6 lessons available

Context preservation, escalation, error propagation, large-codebase exploration, and provenance.

  1. 5.1Manage conversation context to preserve critical information6 questions
  2. 5.2Design effective escalation and ambiguity resolution patterns6 questions
  3. 5.3Implement error propagation strategies across multi-agent systems6 questions
  4. 5.4Manage context effectively in large codebase exploration6 questions
  5. 5.5Design human review workflows and confidence calibration6 questions
  6. 5.6Preserve information provenance and handle uncertainty in multi-agent systems6 questions

Practice Domain 5

Quick reference

Instructional diagrams

Open any diagram in its lesson.

  1. Task 1.1Agent loop and stop-reason routing
  2. Task 1.2Coordinator and specialist orchestration
  3. Task 2.1Tool contract selection
  4. Task 2.2Tool-result failure routing
  5. Task 3.1Instruction scope resolution
  6. Task 3.3Glob pattern reach
  7. Task 4.3Structured-output validation
  8. Task 4.4Evaluation feedback loop
  9. Task 5.1Context placement
  10. Task 5.3Failure propagation
  11. Task 5.6Claim-level provenance