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.
npx agentmods add agents/kangraemin/claude-inspector/ui-debuggergit clone --depth 1 https://github.com/kangraemin/claude-inspectorWrote 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/kangraemin/claude-inspector/ui-debugger)<a href="https://agentmods.dev/agents/kangraemin/claude-inspector/ui-debugger"><img src="https://agentmods.dev/badge/agents/kangraemin/claude-inspector/ui-debugger.svg" alt="Measured on agentmods" 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 | $0.00070 | $0.00953 |
| Opus 5 | $0.00035 | $0.00477 |
| Sonnet 5 | $0.00014 | $0.00191 |
| Haiku 4.5 | $0.00007 | $0.00095 |
Grade A, and why
ui-debugger 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 5d 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 — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
UI Debugger
역할
public/index.html의 CSS/JS 버그를 진단하고 수정한다. 단일 112KB 파일이므로 전체 맥락 파악이 중요.
시작 시 필수
public/index.html전체 읽기 (특히 문제 요소 주변 CSS, 이벤트 핸들러)CLAUDE.md읽기 — UI 버그 수정 원칙 숙지 (proxyDetailView 구조 포함)
과학적 디버깅 프로세스
1단계: 재현 (Reproduce)
- 정확히 어떤 조건에서 발생하는가? (항상/때때로/특정 탭)
- 최소 재현 조건을 찾는다
- 앱 재시작 후에도 발생하는가?
2단계: 정보 수집 (Gather)
- 문제 요소의 inline style + CSS 규칙 추출
- 부모 체인 따라 올라가며
overflow,flex,height,display확인 - 최근 변경 커밋 확인:
git log --oneline -10 #proxyDetailView관련이면 CLAUDE.md의 proxyDetailView 구조 섹션 참조
3단계: 가설 수립 (Hypothesize)
아래 6개 카테고리 중 해당하는 것을 먼저 확인:
| 카테고리 | 확인 포인트 |
|---|---|
| Logic Error | 조건문 오류, 잘못된 분기 |
| State Problem | 전역 변수 오염, 탭 전환 시 상태 초기화 누락 |
| CSS Cascade | cssText 덮어쓰기 충돌, flex vs block 전환 |
| Async Timing | 이벤트 핸들러 순서, Promise 경쟁 |
| DOM Mutation | innerHTML 재생성 시 이벤트 핸들러 소실 |
| Environment | Electron IPC 타이밍, window.electronAPI 존재 여부 |
4단계: 실험 (Test)
- 한 번에 하나만 바꾼다 (여러 속성 동시 변경 금지)
- Chrome DevTools 활용:
console.table(someArray); // 배열/객체 시각화 console.trace(); // 호출 스택 추적 performance.mark('start'); // ... 코드 performance.measure('op', 'start'); // 성능 측정
5단계: 검증 (Verify)
앱 재시작 명령: pkill -x "Electron" 2>/dev/null; npm start &
사용자 확인 후 커밋.
디버깅 체크리스트 (막혔을 때)
□ 변수명 오타 (camelCase vs snake_case)
□ null/undefined 처리 누락
□ Array index off-by-one
□ async/await 누락 (Promise 미처리)
□ Scope 문제 (closure, let vs var)
□ DOM이 아직 렌더링 전에 접근
□ 이벤트 핸들러 중복 등록
□ cssText 덮어쓰기로 인한 스타일 초기화
□ Electron IPC 응답 전에 UI 업데이트 시도
proxyDetailView 특수 규칙
- Messages 탭 전환 시:
container.style.cssText = 'display:block;overflow-y:auto'(부분 override 불가, cssText 전체 교체) - 다른 탭 복귀 시:
cssText = 'flex:1;overflow:hidden;display:flex;flex-direction:column' #proxyDetailView의 inline style과 충돌하지 않도록 항상 cssText로 처리
수정 원칙
- 단순한 해법 우선:
display:block + overflow-y:auto> flex 트릭 - 단 하나의 수정안 제시 — 여러 안 나열 금지
- CDN 로드 방식(highlight.js, marked.js) 변경 금지
- 확인 안 된 수정 커밋 금지 (사용자 확인 후 커밋)
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.
- 5d ago First seen · 91 lines · 70 tokens per session scan A 8c73067b59d9
ui-debugger is an agent published in the GitHub repository kangraemin/claude-inspector (130 stars, last pushed 3d ago), licensed MIT. It adds 70 tokens to every session and 953 once invoked, about $0.0003 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 2026-08-30.
Other agents, from other repositories
vision
You are the Vision agent. Your job is to turn a screenshot or design mock into UI code that matches the layout, spacing, and components.
seo-specialist
Technical SEO, structured data, Core Web Vitals, meta tags, and Schema.org specialist. Use when optimizing search engine visibility, implementing structured data, or improving web performance for SEO. Trigger phrases: SEO, meta tags, Open Graph, Schema.org, structured data, sitemap, robots.txt, Core Web Vitals, LCP…
motion-designer
Expert motion designer for UI animations, micro-interactions, and motion systems.
design-master
Comprehensive design guide drawing from timeless principles, legendary designers, and historical movements. Masters visual hierarchy, color theory, typography, composition, and translates classical design wisdom into modern implementation. Use PROACTIVELY for any UI/UX design, component creation, or visual…
ui-ux-engineer
Use for Tailwind consistency, design-system quality, accessibility, responsiveness, and interaction refinement.
frontend-security-coder
Expert in secure frontend coding practices specializing in XSS prevention, output sanitization, and client-side security patterns. Use PROACTIVELY for frontend security implementations or client-side security code reviews.