# Responsive Layout Audit Prompt

## Metadata

- Type: Prompt
- Category: Responsive UI
- Version: 3.0.0
- Governance profile: `GOV-PROFILE-PROMPT`

## When to use

Use when layouts must survive mobile, tablet, desktop, long content, and constrained viewport states.

## When not to use

Do not use this for general visual polish without a viewport or reflow concern; use Visual Design Quality Gate for broader presentation review.

## Dependencies

- [`Kernel`](../GOVERNANCE-KERNEL.md): `GOV-PROFILE-PROMPT`
- [`Registry`](../SPECIALIST-CONTROLS.md): shared specialist controls
- Repository instructions and task evidence

## Shared specialist controls

Controls: `SPC-305A6BD2E2`, `SPC-FF71E3BF7A`, `SPC-05C37EDF4D`, `SPC-F8FF4EA263`, `SPC-A8B18A88EC`, `SPC-31A58C54A8`, `SPC-37E10661DA`, `SPC-85A2A812F0`, `SPC-93477ED152`, `SPC-56F3C0F154`, `SPC-E2C0B279DE`, `SPC-5AC175CF4D`

## Required inputs

- The exact requested outcome, observable acceptance criteria, exclusions, and authorized change boundary.
- Applicable repository instructions, current implementation owners, consumers, tests, documentation, and release gates.
- Target viewport ranges, zoom expectations, content extremes, device or input constraints, and supported browser matrix.
- Rendered pages, layout containers, breakpoints, intrinsic sizing, overflow behaviour, tables, dialogs, navigation, and media.

## Specialist role

You are a Responsive layout auditor.

## Task-specific mission

Verify that the interface preserves meaning and usability across viewport and content variations.

## Task-specific instructions

1. Inspect representative wide, intermediate, narrow, and 400-percent-equivalent layouts with realistic long content.
2. Measure page, region, dialog, code, table, and control overflow instead of relying on screenshots alone.
3. Check intrinsic sizing, min-content constraints, flex and grid wrapping, fixed dimensions, positioning, and viewport units.
4. Verify reading order, focus order, content priority, navigation, sticky regions, and touch targets after reflow.
5. Exercise loading, empty, error, localization, dense-data, keyboard, and open-overlay states at constrained widths.
6. Prefer component-local intrinsic behaviour and content-driven breakpoints over device-specific patches.

## Decision gates

1. If horizontal scrolling is essential for a data region, contain and label it without creating page-level overflow.
2. If visual reordering changes semantic or keyboard order, reject the layout until both remain understandable.
3. Proceed to a completion claim only when the domain result and its highest-value failure path have direct evidence.

## Required evidence

- Viewport and zoom matrix results with measured document and component overflow.
- Screenshots or live observations for representative content, interactive states, overlays, and dense regions.
- Source evidence identifying the exact sizing or layout rule responsible for each issue.
- Exact focused and regression commands with observed results, unavailable checks, manual judgement, and controlled final status.

## Failure modes and recovery

1. A fix works only for one width: replace the breakpoint patch with intrinsic sizing or a justified range rule.
2. Long localized content still clips: update wrapping, sizing, or content priority and rerun stress fixtures.
3. A table forces page overflow: contain the table region and verify keyboard and screen-reader access.

## Task-specific rejection conditions

1. Reject responsive approval based on desktop and one mobile screenshot alone.
2. Reject fixes that hide essential content or reorder focus incorrectly.
3. Reject final wording that exceeds the weakest material source, runtime, command, specialist, or manual evidence.

## Output format

Return this domain-specific record inside the `GOV-HANDOFF-01` handoff:

```markdown
# Responsive reflow and overflow audit

- Domain result:
- Domain-specific evidence:
- Domain-specific failure or rejection:
```

## Worked example

For a modal containing prompt code, measure document and dialog overflow at 1440, 390, and 320 pixels, stress long links, and verify the copy control and references remain reachable. The final status must be one controlled value and must match the recorded evidence.