lazy-looplisted

# Lazy Loop Skill **Write the least code that works, then say the least about it.** **Default state: OFF.** Both layers activate only when `/lazy <level>` is run or a trigger phrase above is used; activation without a level = `full`. `/lazy off` deactivates both. When off, code and prose are normal. **Deference: keel-v2 is the arbiter.** Its /review tripwires, design conformance, and a11y requirements outrank every laziness rung and every compression level. Where this file and keel-v2 disagree, keel-v2 wins. Two independent axes, applied in order every active turn: 1. **CODE layer -> ponytail.** Before writing any code, climb the laziness ladder. 2. **PROSE layer -> caveman.** After the code is decided, compress the explanation. They never conflict: ponytail governs *what gets written*, caveman governs *how it's described*. Both keep hard safety guarantees. --- ## Layer 1 - ponytail (CODE) Before writing code, stop at the FIRST rung that holds: ``` 1. Does this need to exist? -> no: skip it (YAGNI) 2. Already in this codebase? -> reuse it, don't rewrite 3. Stdlib does it? -> use it 4. Native platform feature? -> use it 5. Installed dependency? -> use it 6. One line? -> one line 7. Only then: the minimum that works ``` Run the ladder AFTER understanding the problem - read the code the change touches, trace the real flow, then pick a rung. Lazy about the solution, never about reading. **Never on the chopping block:** trust-bound