跳到正文
cccskills
免费在 GitHub 公开

browser-qa

Run real browser QA with Playwright. Use when testing a frontend feature, verifying UI before PR, smoke testing after deploy, or investigating reported visual bugs.

中文简介正在准备中。当前显示原文说明。

集成工具Playwright
查看安装方法

包含的文件(2)

  • SKILL.md3.4 KB
  • playwright-snippets.md2.3 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

Browser QA Skill

Test web apps with a real browser. Find bugs users would find.

When to Use

  • After building a frontend feature
  • Before creating a PR for UI changes
  • When user reports "something looks broken"
  • Smoke testing after deploy

What This Skill Does

  1. Launches Playwright with real Chromium
  2. Hits every discoverable page in the app
  3. Checks for console errors, network failures, broken assets
  4. Tests interactive elements (forms, buttons, links)
  5. Verifies mobile responsiveness
  6. Produces a structured QA report

QA Check Categories

1. Page Health

Every page gets these checks:

  • HTTP status is 2xx
  • No JavaScript console errors
  • No failed network requests (4xx/5xx)
  • No missing images or assets
  • Page loads in under 3 seconds

2. Layout & Responsive

  • Desktop (1280px): no horizontal scroll, no overlapping elements
  • Mobile (375px): no horizontal scroll, text is readable, buttons are tappable
  • No content clipped or hidden unintentionally

3. Interactive Elements

  • Buttons: click → something happens (no dead buttons)
  • Links: click → navigates to valid page (no 404s)
  • Forms: submit empty → validation shows. Submit valid → success feedback.
  • Dropdowns/modals: open → content visible. Close → content hidden.

4. Navigation Flow

  • Every link in nav goes somewhere valid
  • Back button works as expected
  • Breadcrumbs (if present) are accurate
  • Auth redirects work (protected page → login → redirect back)

5. Data Display

  • Lists show data (not empty state when data exists)
  • Pagination works (if present)
  • Search/filter updates results
  • Loading states show during data fetch
  • Error states show when API fails

Playwright Patterns & Report Template

See playwright-snippets.md for ready-to-use Playwright code (page tests, mobile viewport, form testing, screenshot on failure) and the QA report template.

Gotchas

  • networkidle waits forever on apps with websockets or polling. Use domcontentloaded or a specific element selector instead if the app has live connections.
  • page.click() on invisible elements passes silently. Always verify the element is visible before interacting. Use await expect(locator).toBeVisible() first.
  • Mobile viewport test without touch simulation misses real bugs. Set hasTouch: true in the browser context, not just the viewport size.
  • Console warnings are not console errors. Don't report React hydration warnings or deprecation notices as bugs. Only console.error and failed network requests are bugs.
  • Screenshots on CI look different than local. Font rendering, antialiasing, and DPI differ between macOS and Linux. Use visual comparison thresholds, not pixel-perfect matching.

Rules

  • Real browser only. Playwright + Chromium. No HTTP-only testing.
  • Headless by default. Don't open visible windows unless asked.
  • Test both viewports. Desktop (1280px) and mobile (375px).
  • Console errors = bugs. Always report them.
  • One-shot mode. Launch browser, test, close. No daemon.
  • Screenshot failures. Take a screenshot when something breaks.
  • Don't fix without asking. Report first, then offer fixes.

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

Creates RPC-style endpoint following layered architecture (Controller → Manager → Repository). Use when creating new API endpoints or CRUD operations.

中文简介正在准备中。当前显示原文说明。

c0x12c/ai-toolkit1082026年6月18日 更新

Write blog posts, guides, tutorials, and long-form content. Sounds like a real person, not AI. Use when the user wants polished written content.

中文简介正在准备中。当前显示原文说明。

c0x12c/ai-toolkit1082026年6月18日 更新

Design RPC-style APIs with layered architecture (Controller → Manager → Repository). Use when creating new API endpoints, designing API contracts, or reviewing API patterns.

中文简介正在准备中。当前显示原文说明。

c0x12c/ai-toolkit1082026年6月18日 更新

Run a structured brainstorm session for startup ideas. Takes a theme or problem and generates ideas with quick gut-checks. Use when the user wants to explore a space or generate new ideas.

中文简介正在准备中。当前显示原文说明。

c0x12c/ai-toolkit1082026年6月18日 更新

CI/CD pipeline patterns for GitHub Actions, PR automation, and deployment workflows. Use when setting up CI, fixing broken pipelines, automating PR checks, or configuring deployment.

中文简介正在准备中。当前显示原文说明。

c0x12c/ai-toolkit1082026年6月18日 更新

Deep competitor analysis. Tear apart a specific competitor or compare multiple competitors. Use when the user names a competitor or asks "who else is doing this?"

中文简介正在准备中。当前显示原文说明。

c0x12c/ai-toolkit1082026年6月18日 更新

查看 c0x12c 的全部技能

报告此技能的问题