Scenario labs

A live exam presents 4 of these 6 labs — 15 possible draws, about 15 items each. Every item on this site is written against a published task statement, so the labs are how you decide what to prepare, not a prediction of what you will see.

CCAR-F 60 items · 120 min 4 of 6 labs per exam 15 possible draws 120 items in the bank

Ordered by the share of the blueprint each lab spans

  • S4 Developer Productivity with Claude

    spans 65%
    • D1 Agentic Architecture & Orchestration 27%
    • D2 Tool Design & MCP Integration 18%
    • D3 Claude Code Configuration & Workflows 20%

    An Agent SDK productivity agent helping engineers explore unfamiliar codebases, understand legacy systems, generate boilerplate, and automate repetitive tasks, using the built-in tools (Read, Write, Bash, Grep, Glob) and integrating MCP servers.

    9 statements in play 20 items in the bank

  • S1 Customer Support Resolution Agent

    spans 60%
    • D1 Agentic Architecture & Orchestration 27%
    • D2 Tool Design & MCP Integration 18%
    • D5 Context Management & Reliability 15%

    An Agent SDK support agent handling returns, billing disputes, and account issues through custom MCP tools (get_customer, lookup_order, process_refund, escalate_to_human), targeting 80%+ first-contact resolution while knowing when to escalate.

    8 statements in play 20 items in the bank

  • S3 Multi-Agent Research System

    spans 60%
    • D1 Agentic Architecture & Orchestration 27%
    • D2 Tool Design & MCP Integration 18%
    • D5 Context Management & Reliability 15%

    A coordinator agent delegating to specialized subagents — web search, document analysis, synthesis, report generation — producing comprehensive cited reports.

    8 statements in play 20 items in the bank

  • S5 Claude Code for Continuous Integration

    spans 40%
    • D3 Claude Code Configuration & Workflows 20%
    • D4 Prompt Engineering & Structured Output 20%

    Claude Code inside a CI/CD pipeline running automated code reviews, generating test cases, and providing pull-request feedback, with prompts designed to be actionable and to minimize false positives.

    5 statements in play 20 items in the bank

  • S2 Code Generation with Claude Code

    spans 35%
    • D3 Claude Code Configuration & Workflows 20%
    • D5 Context Management & Reliability 15%

    Claude Code for generation, refactoring, debugging, and documentation, integrated into the development workflow with custom slash commands and CLAUDE.md configuration, including when to use plan mode versus direct execution.

    7 statements in play 20 items in the bank

  • S6 Structured Data Extraction

    spans 35%
    • D4 Prompt Engineering & Structured Output 20%
    • D5 Context Management & Reliability 15%

    Extracting information from unstructured documents, validating output against JSON schemas, maintaining high accuracy, handling edge cases gracefully, and integrating with downstream systems.

    7 statements in play 20 items in the bank

Which labs reach each domain

A lab is worth building for the domains it is the only route to. Expected items are the blueprint weight applied to a 60-item exam — the exam's own arithmetic, not a forecast of your score.

Domain Weight Expected items Statements Labs that reach it
D1 Agentic Architecture & Orchestration 27% 16.2 7 S1 S3 S4
D2 Tool Design & MCP Integration 18% 10.8 5 S1 S3 S4
D3 Claude Code Configuration & Workflows 20% 12.0 6 S2 S4 S5
D4 Prompt Engineering & Structured Output 20% 12.0 6 S5 S6
D5 Context Management & Reliability 15% 9.0 6 S1 S2 S3 S6