| Day | Topic | Status |
|---|---|---|
| D1 | The agentic loop and what Claude can actually read (/context) | Published 2026-08-05 |
| D2 | The context window & prompt caching | Published 2026-08-05 |
| D3 | Permission modes & sandboxing | Published 2026-08-05 |
| D4 | CLAUDE.md & rules | Published 2026-08-05 |
| D5 | Plan mode and explore→plan→code | Published 2026-08-05 |
| D6 | Checkpoints, rewind and session management | Published 2026-08-05 |
| D7 | Skills | Published 2026-08-05 |
| D8 | Hooks | Published 2026-08-05 |
| D9 | MCP | Published 2026-08-05 |
| D10 | Subagents | Published 2026-08-05 |
| D11 | Worktrees & agent view | Published 2026-08-05 |
| D12 | Agent teams & dynamic workflows | Published 2026-08-05 |
| D13 | Headless & CI | Published 2026-08-05 |
| D14 | Plugins: packaging & distribution | Published 2026-08-05 |
How the syllabus runs
When the official docs change, published lessons are revised in step (the lesson's "Last updated" date moves). With all 14 lessons out, the course is in continuous-revision mode: each day one lesson gets deepened or re-verified against the latest docs, and new advanced topics are appended to the syllabus.