Status: Phase 1 stub. Matures in Phase 5 per §8.1.

Statement

Diagnostic surfaces are for operators, not for end users. They prioritise inspectability, density, and protocol fidelity over polish. They use the 8-color diagnostic palette (F-3), the four diagnostic layout templates (F-4), and <OperatorBar> as their topbar. Their copy speaks in protocol terms (epoch, KP, gift-wrap, ratchet) without softening for a non-technical audience.

Why

The audit found diagnostic routes designed as if they were product surfaces — friendly empty states, narrative copy, brand-gradient CTAs, semantic-palette colors. Every accommodation toward the imagined "casual user" stripped utility from the operator who actually uses the route. An MLS-epoch-sync screen is not a marketing surface; it is an instrument. Treating it like a marketing surface is a category error.

Conversely, refusing to design diagnostics at all (the prior state) produced 12 ad-hoc routes, each invented from scratch, with the visual incoherence the audit catalogued. D-1 says diagnostics get a different design system, not no design system.

How

Topbar is always <OperatorBar> — never the product header
Layout is one of four templates (F-4): hub / gate / content / compare
Palette is 8-color categorical (F-3); semantic palette is forbidden except for error on a genuine error state
Copy uses protocol vocabulary; no "we're sorry" or "something went wrong" — exact failure mode named
Density is high; whitespace is for legibility, not breathing room

Phase 5 deepens this stub with per-route citations, the operator-vs-product decision tree, and audit screenshots showing the category-error counter-examples.