prototypelisted

# Prototype You own the design decision, not the code. The prototype is a throwaway instrument — the real build follows afterward. This is the one place the usual bars invert: **speed over polish, code quality does not matter, no planning.** The rigor is in picking the right design cheaply. ## Steps (paste as todos) 1. **Scope the decision** the prototype exists to make — which layout, interaction, or (for an empirical fork) which behavior/timing/approach. No decision to make → no prototype; just build the thing. 2. **Gather references** when the design space is open — prior art, a quick moodboard of options; let the user pick directions before building. Skip when the direction is set. 3. **Build throwaway in an isolated scratch dir,** separate from production source. Lightest stack that renders the idea (vanilla HTML/CSS/JS, CDN deps) or the smallest script that exercises the question. No production framework, no tests, no abstractions. 4. **Put alternatives behind one switcher** (buttons / a keypress), each variant labeled, so they're compared side by side. 5. **Verify by observation, not assertion** — screenshot each visual variant; log the output/timing for a behavioral one. The observation *is* the test here. 6. **Present variants, tradeoffs, and a recommendation.** The output is the decision plus the throwaway artifact — not shippable code. Hand the chosen direction to the real build. ## Guardrails - Never ship prototype code. Say plainly it