Loop Engineering

• Primitive selection over loop-by-default. Cleanly separate the four axes: when a single run is enough, when a monitor fits better, and when a real loop is the right tool.
• Done-conditions that prove rather than claim. Understand the judgment gap: /goal does not judge, it only reads. Write conditions that check a real artefact, and set a hard ceiling against endless continuation.
• Deterministic gate instead of model vote. Exit code, test, typecheck, and lint as hard truth per turn. Stop-hook as a blocker, TDD as the foundation. The part quick demos regularly skip.
• Failure patterns as a map. Silent failures, polling waste, runaway loops, context reset: the patterns usually learned through expensive personal mistakes, shown here as a map before they hit you.
• Token economics and safe operation. Cache TTL, cadence banding, kill switches, and read-only-first inside the loop body. The unsexy part that keeps you from real damage.

Structure: 7 modules, 16 lessons

• Module 1: Primitive-Selection Framework4 lessons
• Module 2: Done-Conditions and the Judgment Gap2 lessons
• Module 3: Verification Gates2 lessons
• Module 4: Failure Patterns and Observability2 lessons
• Module 5: Token and Cache Economics2 lessons
• Module 6: Kill Switches and Safe Operation2 lessons
• Module 7: Capstone, do not rebuild native primitives2 lessons

Every module comes from real runs: the places where loops fail silently and the gates that came out of them. No syntax tables copied from the docs, only practice, verification, and operations.

Your access is unlocked immediately after payment: all lessons directly in the browser, in your account. Because this is digital content that starts right away, you confirm immediate access during checkout. Details are in the withdrawal information.