AgentSkillsCN

verify-ui

利用Claude-in-Chrome浏览器自动化工具,对前端UI进行可视化验证,截取桌面端与移动端视口下的所有页面截图。

SKILL.md
--- frontmatter
name: verify-ui
description: Visual verification of frontend UI using Claude-in-Chrome browser automation. Takes screenshots of all screens at desktop and mobile viewports.
disable-model-invocation: true
user-invocable: true

Visual UI Verification

Use Claude-in-Chrome browser automation to visually verify the FloodSafe frontend.

Prerequisites

  • Frontend dev server running on http://localhost:5175
  • Chrome browser open

Steps

1. Setup

  • Load Claude-in-Chrome tools via ToolSearch if not already loaded
  • Call tabs_context_mcp to see current browser state
  • Create a new tab with tabs_create_mcp

2. Desktop Screenshots (1400x900)

Navigate to http://localhost:5175 and screenshot each screen:

  1. Login Screenhttp://localhost:5175 (when logged out)
  2. Home Screen — main dashboard after login
  3. Flood Atlas — map view
  4. Alerts — alerts tab
  5. Profile — profile/settings
  6. Report — report submission flow (first step)

Use read_page to check for accessibility issues on each screen.

3. Mobile Screenshots (375x812)

  • Use resize_window to set viewport to 375x812
  • Re-screenshot Home, Atlas, and Alerts screens
  • Check for overflow, hidden elements, navigation usability

4. Console Check

  • Use read_console_messages to check for errors/warnings
  • Report any console errors found

5. Report

Provide a summary:

  • ✅/❌ Each screen renders correctly
  • ✅/❌ Mobile responsive
  • ✅/❌ Console clean (no errors)
  • List any visual issues found with descriptions