Study session
Learn with the AI tutor, or jump into a curriculum lesson.
Ask anything about the five exam domains. The tutor explains concepts in plain language and points you to the most relevant lesson or drill on the site.
Open it with the “Ask the tutor” button in the bottom-right of any page.
Curriculum lessons
Design multi-step Claude agents: when to use loops vs. orchestrator-worker, how to compose subagents, and how to evaluate agent outputs.
Define tools that Claude uses reliably — schema design, error contracts, MCP servers, and resource exposure.
CLI features, hooks, slash commands, skills, settings.json, IDE plugins, and Agent SDK workflows.
System prompts, structured outputs, JSON mode, multi-shot examples, and prompt caching.
Token budgets, context compaction, retries, observability, and production reliability patterns.
