メニュー
Use when auditing imagery, brand assets, UI screens, or AI product features for visual and product slop. Unlike ai-slop-typography-audit, this excludes type; written-copy slop routes to the digital-research engine.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Use when a game for children or learners needs age-appropriate comprehension, scaffolding, learning-feedback separation, privacy/safety, stopping cues, guardian/teacher needs, accessible play, or non-exploitative engagement. Use onboarding-and-first-run-design for ordinary apps.
Use when a game needs visual pillars, shape language, colour scripting, character/world readability, cultural motif governance, 2D/3D cohesion, VFX hierarchy, an asset bible, style scalability, performance-aware art, or engine handoff. Use illustration-style-and-systems for non-game illustration sets.
Use when tuning a game action through anticipation, response, impact, timing, animation, VFX, audio-visual coordination, camera, hit-stop, particles, haptics, reduced-motion, sensory alternatives, or intensity settings. Use micro-interactions-and-feedback for ordinary UI controls.
Use when designing game HUDs, menus, maps, inventory, dialogue, tutorials, pause/settings, save/load, failure/success, controller focus, touch, couch-distance, or diegetic interfaces. Use ordinary app UI skills for non-game task flows.
Use when a game needs its player fantasy and interaction loop routed into HUD, art direction, feedback, camera, input, accessibility, testing, and engine handoff. Use ordinary app UI skills for non-game software and the four game specialists for focused craft.
Use when designing or auditing an interactive UI for WCAG 2.2 AA, keyboard and focus behaviour, accessible names, screen readers, target size, reflow, authentication, or reduced motion. Use inclusive-and-assistive-design for needs beyond the compliance floor.
| name | visual-product-slop-audit |
| description | Use when auditing imagery, brand assets, UI screens, or AI product features for visual and product slop. Unlike ai-slop-typography-audit, this excludes type; written-copy slop routes to the digital-research engine. |
| metadata | {"portable":true,"category":"00-cross-cutting-ops-qa-a11y","compatible_with":["claude-code","codex"]} |
Acknowledgement: Shared by Peter Bamuhigire, techguypeter.com.
The concern is the typeface/type system — use ai-slop-typography-audit.
The concern is written copy (preambles, buzzword salad, vague attribution) — that is a writing
concern owned by the digital-research-engine (anti-ai-slop); route there.
The question is whether the definition of AI slop, a banned font, or a current visual tell has
changed - use slop-doctrine-refresh-and-research-loop first, then return here for artifact
audit.
| Input | Supplied by | Required? | Why |
|---|---|---|---|
| Assets, screens, or live feature | Requester | yes | Defines the audit surface |
| Audience, channel, and stakes | Product or brand brief | yes | Calibrates severity |
| Current slop taxonomy | Design doctrine | yes | Prevents stale or invented tells |
doctrine/references/ai-slop-taxonomy.md).references/visual-tells-checklist.md, which operationalises
ai-slop-taxonomy.md → Visual & video tells): waxy skin, melted backgrounds, gibberish text,
floating objects, extra fingers/impossible anatomy, warped/misspelled logos, uncanny faces,
mismatched lighting. Current 2026 AI-image tells (the classic six-finger artifact is largely
fixed in late-2025/2026 models, so the tells migrated): plastic "default-render" lighting & creamy
bokeh, texture-too-perfect / uniform-frequency detail, micro-anatomy & accessory errors visible
only on zoom (teeth, earrings, watch faces, fused hair), garbled small text / UI chrome / gauges,
impossible reflections & re-tiling patterns, and the "no-author" style-blend aesthetic — perfect,
even, styleless competence that looks made by nobody. The 2026 rule of thumb: the tell is now the
absence of authored specificity, not a hard anomaly.| Output | Consumer | Evidence / acceptance |
|---|---|---|
| Severity-rated finding register | Designer and owner | Asset location, tell, evidence, and taxonomy mapping |
| Remake/remediate disposition | Production team | Named human-craft alternative for every finding |
| Gate verdict | Pre-launch QA | PASS, CONDITIONAL, or BLOCKED with unresolved risks |
examples/slop-audit-filled.md for a complete severity and disposition register.| Condition | Decision | Wrong-choice failure |
|---|---|---|
| Identity, anatomy, or embedded text is corrupted | Remake or replace and block ship | Visible anomaly damages trust |
| Asset is competent but unauthored and generic | Re-art-direct with a specific point of view | Convergent brand work ships |
| AI feature lacks grounding, verification, or undo | Remove or ground it | Generated output is treated as fact |
| Evidence is ambiguous | Mark conditional and seek source/provenance | Suspicion is presented as proof |
Read and high-resolution visual inspection are required. Review is read-only unless remediation is authorised. Network access is optional for provenance checks; editing or generation must preserve brand and usage rights.
Without original-resolution assets or live-flow access, audit the supplied evidence, mark hidden details unverified, and withhold a pass. Recover by requesting crops, source files, or a recorded walkthrough.
examples/slop-audit-filled.md — a real before/after audit of a Maduuka landing-page hero +
embedded AI chat panel: findings (tagged critical/major/minor) → remake/remediate dispositions
with the human-craft alternative named for each.references/visual-tells-checklist.md — the concrete, tickable image + product/interface tells
(incl. the current 2026 AI-image tells), citing the taxonomy.doctrine/references/ai-slop-taxonomy.md, doctrine/design-doctrine.md (Mission).doctrine/references/living-slop-refresh-protocol.md and
slop-doctrine-refresh-and-research-loop for refreshing changing slop definitions.ai-slop-typography-audit (type); digital-research anti-ai-slop (writing).