Claude Architect Foundations (CCA-F) Exam Prep
Prepare for the Claude Architect Foundations (CCA-F) exam with 37 self-paced lessons in nine modules, covering all five exam domains.
- Agentic architecture and orchestration: the agentic loop, ending it on stop_reason with iteration caps, hub-and-spoke coordinators, task decomposition, subagents, and session resume and fork in the Agent SDK.
- Tool design and MCP: tool descriptions and JSON Schema inputs, MCP tools vs. resources vs. prompts, server discovery and MCP permission rules.
- Claude Code configuration: settings scopes, deny/ask/allow permission rules, permission modes, sandboxing, session commands and GitHub Actions.
- Prompting and structured output: few-shot examples, tool_choice, schema-enforced output, validation-retry loops and fresh-context review passes.
- Context management and reliability: large-context failure modes, scratchpads and external memory, provenance, the Message Batches API and escalation to a human.
- Practice: a 15-item diagnostic, five domain knowledge checks and a timed 38-question practice exam at two minutes per question, with a rationale for every option.
Preview mode. Get a plan to unlock all lessons, quizzes, and exams.
Orientation
doc About the Claude Architect Foundations Certification locked
doc Exam Logistics and Study Plan locked
exam Day-One Diagnostic: Find Your Weak Domains locked
The agentic loop
doc The Agentic Loop locked
doc Tools and Tool Use locked
doc Choosing Models locked
doc stop_reason: How the Loop Communicates locked
doc Ending the Loop Safely: Termination Anti-Patterns locked
Orchestration
doc Code-Driven vs. Model-Driven Decision-Making locked
doc Hub-and-Spoke: The Coordinator Pattern locked
doc Task Decomposition and Its Failure Modes locked
doc Dynamic Selection and Partitioned Research locked
doc Refinement Loops and Coordinator Observability locked
doc Subagent Failure Handling and Handoff Protocols locked
Agent sdk
doc The SDK Landscape: Anthropic SDK vs. Claude Agent SDK locked
doc Agent Definitions locked
doc Subagents and Parallelism locked
doc Sessions, Resume, and Fork locked
quiz Knowledge Check D1: Agentic Architecture & Orchestration locked
Claude code configuration
doc Settings Files and Scopes locked
doc Permission Rules locked
doc Permission Modes locked
doc Sandboxing locked
doc dangerously-skip-permissions locked
doc Session and Context Commands locked
Prompting and structured output
doc Specificity and Few-Shot Prompting locked
doc Goals and Criteria Over Procedures locked
doc Controlling Tool Invocation with tool_choice locked
doc JSON Schema and Structured Output via Tool Inputs locked
doc Validation-Retry Feedback Loops locked
doc Independent Review: Beating Self-Review Bias locked
quiz Knowledge Check D4: Prompt Engineering & Structured Output locked
Context management
doc Large-Context Failure Modes locked
doc External Memory: Findings Files and Scratchpads locked
doc Preserving Provenance in Multi-Source Synthesis locked
doc Batch Processing locked
Mcp
doc MCP Servers and Tool Discovery locked
doc MCP Tools, Resources, and Prompts locked
quiz Knowledge Check D2: Tool Design & MCP Integration locked
Operations
doc Claude Code in GitHub Actions locked
doc Designing Escalation to Humans locked
quiz Knowledge Check D3: Claude Code Configuration & Workflows locked
quiz Knowledge Check D5: Context Management & Reliability locked
Exam readiness
exam CCA-F Practice Exam (timed) locked