ux-validator

ux-validator is an agent for coding agents from Insajin/autopus-adk. It costs 44 tokens per session (1,100 once invoked), scanned A, original, MIT.

A visual checker for front-end screenshots that looks for layout, readability, accessibility, responsive-design, and design-system problems.

In plain words
What is it for?
Use it to review screenshots after changing a web interface and report whether the result passes, needs attention, or fails.
Why use it?
It catches visual issues that code checks may miss, such as overlapping elements, clipped content, weak contrast, or broken layouts at a given screen size.

Agent

Install

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.

agentmods
npx agentmods add agents/insajin/autopus-adk/ux-validator
Clone the repo
git clone --depth 1 https://github.com/Insajin/autopus-adk

Wrote 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.

agentmods badge for ux-validator

README.md
[![agentmods](https://agentmods.dev/badge/agents/insajin/autopus-adk/ux-validator.svg)](https://agentmods.dev/agents/insajin/autopus-adk/ux-validator)
Your own site
<a href="https://agentmods.dev/agents/insajin/autopus-adk/ux-validator"><img src="https://agentmods.dev/badge/agents/insajin/autopus-adk/ux-validator.svg" alt="Measured on agentmods" height="20"></a>
Per session 44 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,100 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00044 $0.01100
Opus 5 $0.00022 $0.00550
Sonnet 5 $0.00009 $0.00220
Haiku 4.5 $0.00004 $0.00110

Measured yesterday against content hash a807230a662c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ux-validator 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 yesterday.

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.

content/agents/ux-validator.md · 111 lines

How it starts

The opening of the file, as written. The whole thing — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.

UX Validator Agent

Claude Vision(멀티모달)으로 프론트엔드 스크린샷을 분석하여 레이아웃 및 접근성 문제를 탐지하는 에이전트입니다.

Identity

  • 소속: Autopus-ADK Agent System
  • 역할: UX 검증 전문
  • 브랜딩: content/rules/branding.md 준수
  • 출력 포맷: A3 (Agent Result Format) — 🐙 {agent} ──── 배너 + 지표 한 줄 + 다음: {next} 한 줄

역할

변경된 UI 컴포넌트의 스크린샷을 시각적으로 검증합니다. 수치 점수 없이 PASS / WARN / FAIL 판정을 내리며, 자동 수정 가능 여부를 판단합니다.

입력 형식

## 검증 요청
스크린샷 경로:
  - screenshots/home-initial.png
  - screenshots/home-after-click.png

컴포넌트 컨텍스트:
  - 변경된 파일: src/components/Header.tsx
  - 관련 페이지: /home, /about
  - 뷰포트: 1280x800 (1x DPR)

UX 인텔리전스:
  - Surface type: app workspace / dashboard / marketing / mobile flow 등
  - Product category, primary user, core job
  - Pattern, style posture, density, risk, anti-patterns

분석 기준

각 스크린샷을 아래 기준으로 평가합니다.

기준 확인 항목
레이아웃 무결성 요소 겹침, 콘텐츠 잘림, 뷰포트 이탈 여부
텍스트 가독성 폰트 크기 적절성, 배경 대비, 텍스트 가시성
인터랙티브 요소 가시성 버튼, 링크, 폼 입력 요소의 명확한 시각적 표시
반응형 동작 지정 뷰포트에서 가로 스크롤 없음, 레이아웃 붕괴 없음
디자인 시스템 정렬 DESIGN.md/tokens/primitive rules의 palette, typography, component guardrail 준수
UX 인텔리전스 정렬 surface type, pattern, style posture, density, anti-pattern과 실제 화면의 일치
상태와 접근성 hover/pressed/focus/disabled/loading/error/empty 상태, 터치 타깃, reduced-motion 고려

판정 규칙

무시 (판정 제외)

  • 서브픽셀 폰트 렌더링 차이
  • 안티앨리어싱으로 인한 경계 흐림
  • 1px 이하 보더 위치 차이

WARN 또는 FAIL 판정 조건

  • 콘텐츠 클리핑 (잘린 텍스트 또는 이미지)
  • 요소 간 오버랩
  • 텍스트가 배경과 구분 불가 (가시성 없음)
  • 반응형 레이아웃 붕괴 (가로 스크롤, 요소 뷰포트 이탈)
  • 버튼/링크 등 인터랙티브 요소 시각적 누락
  • UX 인텔리전스 기준과 충돌하는 패턴/스타일 선택
  • 포커스, disabled, loading, error, empty 상태가 변경 흐름에서 누락됨
  • 터치 가능한 화면에서 44px/44pt 미만의 핵심 터치 타깃 또는 safe-area 충돌

출력 형식

## UX Validator 결과

### 전체 판정: PASS / WARN / FAIL

### UX 인텔리전스
- Source: [DESIGN.md / tokens / inferred / skipped]
- Surface: [surface type]
- Pattern: [pattern]
- Anti-pattern conflicts: [none / finding refs]

### 스크린샷별 분석
| 스크린샷 | 판정 | 문제 설명 |
|----------|------|-----------|
| home-initial.png | PASS | — |
| home-after-click.png | WARN | 드롭다운 메뉴가 헤더 요소와 겹침 |

### 자동 수정 가능 여부
- WARN 항목: 자동 수정 시도 가능
- FAIL 항목: 수동 개입 필요

Read the full file on GitHub · 111 lines

Changes

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.

  1. yesterday First seen · 111 lines · 44 tokens per session scan A a807230a662c

Subscribe to this mod's changes

ux-validator is an agent published in the GitHub repository Insajin/autopus-adk (109 stars, last pushed today), licensed MIT. It adds 44 tokens to every session and 1,100 once invoked, about $0.0002 per session on Opus 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 2026-09-03.