CCA-F › Domain 1

Domain 1 — Agentic Architecture & Orchestration

27% of the exam · the biggest & hardest domain · 44 deep-dive subtopics

Exam traps and practice for this domain

On the exam every option in this domain is a sensible practice; these traps decide most questions. Each trap page has a simulator and exam-style practice.

Practice: 68 plain-style questions on D1, each with “Explain simply” and “Think it through”. New here? Start here.

🗺️ Support agent mind maps: scenario 1 on one printable page ›
Six maps: when the loop stops, prompt or hook, cleaning results, escalation, several matches, keeping exact facts.
🗺️ Multi-agent mind maps: the research scenario on one printable page ›
Which scenarios ask about it, then six maps: coverage gaps, subagent context, fixed or adaptive, tools per agent, failures, sources.

This is the curriculum hub for Domain 1. Each subtopic below is a self-contained, architect-level page: concept → diagram → runnable code → trade-offs → exam traps → practice questions. Work top-to-bottom, or jump to any cluster.

Runnable Python demos live in examples/. Diagrams render via Mermaid.js (needs internet on first load).
🧠 Agentic Design Patterns — the full catalog: every pattern (chaining, routing, parallelization, orchestrator-workers, evaluator-optimizer, the autonomous loop, ReAct, reflection, verifier, HITL, memory, RAG, guardrails) with what · why · how in Claude (client/server) · example · gotcha.

DIAGRAM The Domain 1 learning roadmap

How the seven clusters build on each other — from the raw agentic loop, up to multi-agent orchestration, then the safety, reliability, and cost levers that make it production-ready.

flowchart TD START(["🎯 Domain 1 · 27% of the exam
44 subtopics · 7 clusters"]):::start START --> C1 subgraph BUILD ["① Build the loop"] direction LR C1["1 · Foundations
the agentic loop"] --> C2["2 · Stop Reasons
turn control"] --> C3["3 · Tool Use
request → result cycle"] end C3 --> C4["4 · Multi-Agent
orchestration"]:::scale C4 --> C5 subgraph SAFE ["② Make it safe & reliable"] direction LR C5["5 · Autonomy
& guardrails"] --> C6["6 · Reliability
failure & recovery"] end C6 --> C7["7 · Performance
state & cost levers"]:::perf C7 --> GOAL(["✅ Exam-ready architect"]):::goal class C1,C2,C3 build; class C5,C6 safe; classDef start fill:#e2d4f0,stroke:#764ba2,stroke-width:2px,color:#3d1a5b; classDef goal fill:#d4edda,stroke:#28a745,stroke-width:2px,color:#155724; classDef build fill:#d1ecf1,stroke:#17a2b8,color:#0c5460; classDef scale fill:#e2d4f0,stroke:#764ba2,color:#3d1a5b; classDef safe fill:#fff3cd,stroke:#d39e00,color:#7a5b00; classDef perf fill:#d4edda,stroke:#28a745,color:#155724;

Cluster 1 · Foundations of the Agentic Loop

Cluster 2 · Stop Reasons & Turn Control

Cluster 3 · Tool Use within Orchestration

Cluster 4 · Multi-Agent Orchestration

Cluster 5 · Autonomy & Guardrails

Cluster 6 · Reliability, Failure & Recovery

Cluster 7 · Performance, State & Cost Levers

Also see the one-page visual overview & flashcards in overview.html · runnable code in examples/.