Chapter

TL;DR — if you do only five things

1. Be specific and positive. Name files, constraints, and the pattern to follow; say what to do, not what to avoid.
2. Give the agent an executable Definition of Done. Tests/lint/typecheck as commands — that's the loop's exit condition.
3. Plan before you edit; slice the work thin. Investigate → approve a plan → one vertical slice at a time.
4. Commit on every green step. Each commit is a checkpoint the loop can revert to.
5. Engineer the environment, not the prompt. CLAUDE.md, Skills, hooks, MCP, CI carry the intelligence.