메뉴
Live browser verification using playwright-mcp — navigate a running app, screenshot key flows, check accessibility snapshots, verify against use cases. Works with any LLM (no vision required). Use after implementation or for regression checks.
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
직접 명령은 검토 Prompt를 거치지 않습니다. 실행하기 전에 소스를 확인하세요.
npx skills add https://github.com/bpmforge/attest-claude --skill ui-verifier명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.
로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.
Gauntlet loop — multi-agent quality harness: a lead sets a real reference bar (a named product, test suite, or baseline to match or beat), splits the goal into gradeable units, builders produce artifacts in clean context, and blind fresh-per-round critics grade each artifact against the bar with evidence; failures loop until every unit passes, two rounds stall, or budget runs out. NOT /challenge (verifies factual claims); NOT /review (one-pass verdict) — this rebuilds until the work beats something real.
Claude-Design-style visual iteration loop — render the running UI, screenshot at 375/768/1440, critique against docs/design/tokens.json and design principles, apply fixes, re-capture until clean (cap 3 iterations). Also extracts a token baseline from an existing codebase (--sync) and audits real logged-in browsers (--real). NOT a one-pass review — use /ux --review for findings-only; NOT functional conformance — use /ui-verify.
Project intelligence steward — audits CLAUDE.md / AGENTS.md alignment with actual codebase, captures session learnings, updates project docs. Use after major sessions or when docs feel stale.
SOC 직업 분류 기준
SKILL.md 표시 중
| name | UI Verifier |
| trigger | /ui-verify |
| description | Live browser verification using playwright-mcp — navigate a running app, screenshot key flows, check accessibility snapshots, verify against use cases. Works with any LLM (no vision required). Use after implementation or for regression checks. |
| context | fork |
| agent | ui-verifier |
| arguments | [{"name":"url","description":"Base URL of the running app (e.g. \"http://localhost:3000\"). Omit if obvious from context.","required":false},{"name":"--smoke","description":"Quick pass — navigate main routes, screenshot each, check for errors (default)","required":false},{"name":"--use-cases","description":"Verify P0 use cases from docs/testing/USE_CASES.md one by one","required":false},{"name":"--flow","description":"Verify one specific flow end-to-end (e.g. --flow \"user login and dashboard load\")","required":false},{"name":"--regression","description":"Post-change regression check — screenshot key views, flag structural changes","required":false}] |
Triggers the ui-verifier subagent in a forked context.
Live browser verification specialist using playwright-mcp. Navigates a running
application, executes flows, and produces docs/test/UI_VERIFICATION_REPORT.md
with per-flow PASS/FAIL/WARN verdicts and descriptions of what was observed.
Works without vision: Uses accessibility tree snapshots (browser_snapshot())
as the primary verification signal — no vision-capable model required. Screenshots
are taken for visual record and described if vision is available.
Modes:
--smoke (default) — 5 min quick pass over main routes--use-cases — verify every P0 use case in USE_CASES.md--flow "description" — single flow end-to-end--regression — post-change check against last-known stateRequires: playwright-mcp registered (claude mcp list should show playwright).
If not registered, run ./install.sh or claude mcp add playwright -- npx -y @playwright/mcp@latest.