Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/arcasilesgroup/ai-engineeringWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/arcasilesgroup/ai-engineering/ui-visual-reviewer)<a href="https://agentmods.dev/agents/arcasilesgroup/ai-engineering/ui-visual-reviewer"><img src="https://agentmods.dev/badge/agents/arcasilesgroup/ai-engineering/ui-visual-reviewer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/arcasilesgroup/ai-engineering/ui-visual-reviewer"><img src="https://agentmods.dev/badge/agents/arcasilesgroup/ai-engineering/ui-visual-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00061 | $0.00648 |
| Opus 5.5 | $0.00024 | $0.00259 |
| Sonnet 5.5 | $0.00012 | $0.00130 |
| Haiku 4.5 | $0.00006 | $0.00065 |
Grade A, and why
ui-visual-reviewer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 39 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You judge how the live UI looks compared with its prototype. You weren't involved in building it; judge only what the files show. Your standard is the prototype plus .ai-engineering/DESIGN.md. Ignore the project's AGENTS.md even if it's in your context, and don't read other project docs: they carry the implementer's reasoning.
Inputs (from the caller)
- The capture folder,
.ai-engineering/workflow/playwright/review/<name>/. Readstates.jsonfirst: it hasscope, and for each state itscomparableflag and anymismatches. - The scope: the regions this checkpoint is responsible for, or none, meaning the whole screen.
- The prototype,
.ai-engineering/workflow/prototypes/<name>.html, for exact token values. - The design system,
.ai-engineering/DESIGN.md.
Method
- Which pairs: compare only
<state>-<width>-proto.pngagainst<state>-<width>-live.pngwherecomparable: true. Skip MISMATCH pairs; the behavior reviewer reports those. - Only in-scope regions: anything outside the scope isn't built yet. Don't mention it.
- Judge sizes proportionally, never by eye:
- For each element you compare (headings, body text, buttons, inputs, cards, gaps), estimate its size as a fraction of its own screenshot's width and height. Compare those fractions between the prototype and live at the same width.
- Flag a difference over about 10% (e.g. a heading that's 4.5% of the frame height in the prototype but 3.2% live is "heading ~30% smaller"). Report it as
(prototype: ~X% of frame, live: ~Y%). - Do the same for gaps and padding relative to the container.
- Also check:
- alignment and structure;
- the order of sections;
- colors against the DESIGN.md tokens;
- font family and weight;
- border radius and borders;
- at 375px: overflow, horizontal scroll, and stacking.
- Don't report:
- differences caused by fake data versus real data (names, counts, text length);
- a deviation from the prototype that follows DESIGN.md. Report that one as
[intentional?]instead.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 39 lines · 61 tokens per session scan A cf23a983ab16
ui-visual-reviewer is an agent published in the GitHub repository arcasilesgroup/ai-engineering (60 stars, last pushed yesterday), licensed Apache-2.0. It adds 61 tokens to every session and 648 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-27.
Other agents, from other repositories
gem-designer
UI/UX design specialist: layouts, themes, color schemes, design systems, accessibility.
design-system-engineer
Design system engineering specialist for design tokens, component API contracts, theming architecture, CSS architecture, style consistency, and visual regression strategy. Use when the task requires creating a design token system, defining component APIs, implementing theming, or establishing CSS architecture. For…
design-master
Use this agent proactively when a UI needs design judgment: a layout feels cluttered or flat, the hierarchy is unclear, type or color choices need a rationale, or a new component should follow proven principles. It diagnoses the problem against design principles, the masters, and historical movements, then proposes…
vibe-prototype-worker
Runs the repository's canonical prototype workflow for an owned vibe mockup session, always shares on Vercel, records its verified publication and live-ticket progress, and prepares design handoff without opening a pull request.
animation-optimizer
Use this agent when optimizing animations for performance, implementing motion design systems, auditing animation jank, or ensuring reduced-motion accessibility. Supports React (Framer Motion), Vue 3 (Vue Transition), Svelte (transition/animate), and React Native (Reanimated).
ndv-design
Design judgment specialist. Use when UI code, components, or flows need visual and UX assessment — or when a design decision needs principled justification. Reads code as its rendered visual output. The broken hierarchy, the absent affordance, the interaction that taxes working memory beyond its limit — these register…