Handbook
The same five domains as the CCA-F curriculum, written for building — not for passing a test. No exam-trap callouts, just what actually works in production, with real code. If you're studying for the exam, the curriculum is the version tuned for that.
Agentic Loops and Multi-Agent Orchestration
How the model-tool loop actually runs, five patterns for structuring multi-step work, and what breaks when agent systems go to production.
Read →Designing Tools and MCP Integrations
How to write tool schemas Claude calls correctly, wire up the tool_use loop, and decide when MCP is the right integration layer.
Read →Configuring Claude Code for Teams
How to set up CLAUDE.md, permissions, slash commands, skills, hooks, and CI so Claude Code behaves predictably across a team.
Read →Prompt Engineering Patterns That Actually Work
How to split system and user roles, structure prompts with XML, choose few-shot vs chain-of-thought, and force reliable structured output.
Read →Context and Reliability in Production
How to budget context, cache prompts, compact history, and design retries so agents stay fast, cheap, and correct.
Read →