Build the Lever

When the work isn't trivial, build the tool that does it instead of doing it by hand.

Why: Two payoffs. Throughput: a codemod, generator, or script does the work the same way every time and reruns for free. Confidence: the tool is one artifact a reviewer can read and rerun to check the work. Hand-done changes can only be re-verified by redoing them. A deterministic script turns "trust me" into "run this".

Pattern: Default to building the lever. Skip it only when the task is genuinely trivial, a couple of obvious edits you can see at a glance.

• Do the first unit by hand to learn the recipe, then build the tool. Prove it by rerunning it on that unit and diffing against your hand-done version. Make the lever safe to rerun. A reviewer will.
• Codemod or script for edits, generator for repetitive files, a dump-to-sqlite query for analysis, a rerunnable check for verification.
• A deterministic lever beats fan-out. If the tool can process every unit in one pass, run it yourself; don't fan out delegates to hand-apply what a script can do.
• When independent work passes ../david-mode/references/agent-runtime.md, put its shared recipe, verification contract, and write fences in one named file. The parent owns that file and dispatch; workers read it without changing the contract. Create a reusable skill only when the recipe will outlive this run.
• Applying this principle produces a reusable artifact: a codemod, script, generator, or bounded worker recipe. A chat claim alone does not satisfy it.
• Commit the lever when the work outlives the session, so the next run reruns it instead of redoing it.

Balance: The bar is triviality, not repetition. A one-off still earns a lever when the lever is what makes the work checkable. Per the Laziness Protocol, build the smallest script that does or proves the job, never a framework.

Distinct from Encode Lessons in Structure, which makes a recurring instruction a durable guardrail. This is throughput and reviewability on the work in front of you. For scripting the verification itself, see Prove It Works.