Claude Certified Architect - Foundations Practice Test

What this exam covers

The Claude Certified Architect – Foundations certification (exam code CCAR-F) validates that practitioners can make sound architectural and configuration decisions when building production-grade applications with Claude. The exam tests practical judgment — not just conceptual recall — across five weighted domains:

• Agentic Architecture & Orchestration (27%) — agentic loops, coordinator-subagent patterns, parallel subagent execution, hooks, and session management
• Claude Code Configuration & Workflows (20%) — CLAUDE.md hierarchy, .claude/rules/ path-scoped rules, custom slash commands and skills, plan mode vs. direct execution, and CI/CD integration
• Prompt Engineering & Structured Output (20%) — few-shot prompting, explicit review criteria, tool_use with JSON schemas, batch processing with the Message Batches API, and multi-pass review architectures
• Tool Design & MCP Integration (18%) — writing effective tool descriptions, structured error responses, tool distribution across agents, and MCP server configuration
• Context Management & Reliability (15%) — context window optimization, escalation patterns, error propagation, provenance tracking, and human review workflows

Questions are grounded in realistic scenarios — building a customer support resolution agent, designing a multi-agent research pipeline, integrating Claude Code into CI/CD, and extracting structured data from unstructured documents — drawn from a bank of six, with four presented per exam sitting.

The exam consists of 60 items (multiple-choice and multiple-response) delivered in 120 minutes via Pearson VUE (online proctored or test center). The passing score is a scaled score of 720 on a 100–1,000 scale, and the credential is valid for 12 months. The ideal candidate has 6+ months of hands-on experience building with the Claude API, Claude Agent SDK, Claude Code, and Model Context Protocol (MCP), with practical familiarity with both capabilities and production limitations of large language models.