audit-ai-design-slop

Provide a screenshot, rendered page, or source code and request an audit. The review inspects decorative stacking, template repetition, typography clutter, motion theater, and fake proof, then returns prioritized findings with concrete evidence for each.

It detects stacked decorative effects like glow and glass, repeated card and tile templates, filler copy, distracting motion, unverifiable testimonials or metrics, and established failures like low contrast, overflow, and unclear primary actions.

Yes, screenshots and recordings are valid audit inputs alongside rendered pages and source code. Anything not visible in the provided evidence is marked as unknown rather than guessed.

No, the audit defaults to removal and minimal corrections within the existing visual system. Replacement suggestions appear only when deletion would cause a real loss or when the user explicitly requests design swaps.

Findings are ranked P0 through P3: P0 blocks completion or involves deceptive proof, P1 harms comprehension or trust, P2 covers repeated slop weakening hierarchy, and P3 is minor polish. Reports return the five to eight highest-impact items.