Token导航 LogoToken导航

Agent Skill · 研究检索

senior-frontend(高级前端)

用于辅助前端页面、组件、样式和交互逻辑的开发与维护。它适合让 Agent 生成或审查 React、Next.js、Vue、Tailwind、CSS 等相关代码,整理组件结构,或定位布局和性能问题。使用时需要结合项目现有设计系统、路由和构建方式,避免只生成孤立片段;涉及页面改动时,应配合本地预览和构建检查确认视觉效果。

senior-frontend 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为操作浏览器;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 47,060。

总安装 751周安装 31GitHub Stars 1下载 246核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:senior-frontend(高级前端)
来源仓库:https://github.com/pixel-process-ug/superkit-agents
仓库路径:skills/senior-frontend
安装命令:
npx skills add https://github.com/pixel-process-ug/superkit-agents --skill senior-frontend
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/pixel-process-ug/superkit-agents --skill senior-frontend

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

senior-frontend 用于辅助前端页面、组件、样式和交互逻辑的开发与维护,适合生成或审查相关代码。适用于前端开发中的技术指导,可整理组件结构或定位布局问题。使用时需结合项目现有设计系统和构建方式,避免生成孤立片段。涉及页面改动时应配合本地预览确认视觉效果。

适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/senior-frontend

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/pixel-process-ug/superkit-agents --skill senio…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要操作浏览器,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

senior-frontend 怎么安装?

在终端执行 npx skills add https://github.com/pixel-process-ug/superkit-agents --skill senior-frontend。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

senior-frontend 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

senior-frontend 需要什么权限?

权限级别是操作浏览器。该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

senior-frontend 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

senior-frontend 用的人多吗?

公开数据是累计安装 751 次,近一周安装 31 次,下载 246 次,GitHub 1 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

senior-frontend 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
senior-frontend

高级前端 · 本页

7511操作浏览器未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Senior Frontend Engineer

Overview

Deliver production-grade frontend code following a structured three-phase workflow: context discovery, development, and handoff. This skill enforces strict quality standards including atomic design component architecture, comprehensive state management patterns, SSR/SSG/ISR optimization, and mandatory >85% test coverage with Vitest, React Testing Library, and Playwright.

Announce at start: "I'm using the senior-frontend skill for production-grade React/TypeScript development."


Phase 1: Context Discovery

Goal: Understand the existing codebase before writing any code.

Actions

  1. Analyze existing codebase structure and conventions
  2. Identify the tech stack version (React 18/19, Next.js 14/15, TypeScript version)
  3. Review existing component library and design system
  4. Check state management approach already in use
  5. Understand build tooling and CI pipeline
  6. Map existing test infrastructure and coverage

STOP — Do NOT proceed to Phase 2 until:

  • Tech stack versions are identified
  • Existing patterns and conventions are documented
  • Test infrastructure is mapped
  • State management approach is identified

Phase 2: Development

Goal: Implement with strict TypeScript, atomic design, and TDD.

Actions

  1. Design component architecture following atomic design
  2. Implement with TypeScript strict mode
  3. Write tests alongside implementation (TDD when appropriate)
  4. Optimize for performance (bundle size, rendering, loading)
  5. Ensure accessibility compliance

Component Architecture Decision Table (Atomic Design)

LevelDescriptionBusiness LogicExample
AtomsSmallest building blocksNoneButton, Input, Icon, Badge
MoleculesComposed of atomsMinimalFormField, SearchBar, Card
OrganismsComplex with business logicYesDataTable, NavigationBar, CommentThread
TemplatesPage structure without dataLayout onlyDashboardLayout, AuthLayout
PagesTemplates connected to dataData fetchingUsersPage, SettingsPage

Atom Example

interface ButtonProps extends React.ButtonHTMLAttributes<HTMLButtonElement> {
  variant?: 'primary' | 'secondary' | 'ghost' | 'danger';
  size?: 'sm' | 'md' | 'lg';
  isLoading?: boolean;
}

export function Button({ variant = 'primary', size = 'md', isLoading, children, ...props }: ButtonProps) {
  return (
    <button className={cn(buttonVariants({ variant, size }))} disabled={isLoading || props.disabled} {...props}>
      {isLoading ? <Spinner size={size} /> : children}
    </button>
  );
}

State Management Decision Table

State TypeSolutionWhen to Use
Server stateReact Query / TanStack QueryAPI data, caching, sync
Form stateReact Hook Form + ZodForm validation, submission
Global UI stateZustandTheme, sidebar open, modals
Local UI stateuseState / useReducerComponent-specific state
URL statenuqs / useSearchParamsFilters, pagination, tabs
Complex localuseReducerMultiple related state transitions
Shared contextReact ContextTheme, locale, auth (infrequent updates)

SSR / SSG / ISR Decision Table (Next.js App Router)

PatternUse WhenCache Strategy
Static (SSG)Content rarely changesBuild time
ISRContent changes periodicallyRevalidate interval
SSRContent changes per requestNo cache
ClientUser-specific, interactiveBrowser

Server vs Client Component Decision

NeedComponent Type
Direct data fetchingServer (default)
Event handlers (onClick, onChange)Client ('use client')
useState / useReducerClient
useEffect / useLayoutEffectClient
Browser APIs (window, localStorage)Client
Third-party libs using client featuresClient
No interactivity neededServer (default)

STOP — Do NOT proceed to Phase 3 until:

  • Components follow atomic design hierarchy
  • TypeScript strict mode is enabled, no any types
  • Tests are written for all components
  • Accessibility is verified (axe-core)

Phase 3: Handoff

Goal: Verify quality gates and prepare for review.

Actions

  1. Verify test coverage meets >85% threshold
  2. Run full lint and type check
  3. Document complex components with JSDoc/TSDoc
  4. Create Storybook stories for UI components
  5. Performance audit (Lighthouse, bundle analysis)

Performance Checklist

  • Bundle size < 200KB gzipped (initial load)
  • Largest Contentful Paint < 2.5s
  • First Input Delay < 100ms
  • Cumulative Layout Shift < 0.1
  • Images: next/image with proper sizing and formats
  • Fonts: next/font with display swap
  • No layout thrashing (batch DOM reads/writes)
  • Virtualization for lists > 100 items

Coverage Thresholds

{
  "coverageThreshold": {
    "global": {
      "branches": 85,
      "functions": 85,
      "lines": 85,
      "statements": 85
    }
  }
}

STOP — Handoff complete when:

  • Test coverage >85% verified
  • Lint and type check pass with zero errors
  • Performance audit completed
  • Complex components documented

Testing Requirements

Unit Tests (Vitest + React Testing Library)

describe('Button', () => {
  it('renders children', () => {
    render(<Button>Click me</Button>);
    expect(screen.getByRole('button', { name: 'Click me' })).toBeInTheDocument();
  });

  it('shows loading state', () => {
    render(<Button isLoading>Click me</Button>);
    expect(screen.getByRole('button')).toBeDisabled();
  });

  it('calls onClick when clicked', async () => {
    const onClick = vi.fn();
    render(<Button onClick={onClick}>Click me</Button>);
    await userEvent.click(screen.getByRole('button'));
    expect(onClick).toHaveBeenCalledOnce();
  });
});

Integration Tests

  • Component compositions (form submission flow)
  • Data fetching with MSW (Mock Service Worker)
  • Routing and navigation
  • Error boundaries and fallbacks

E2E Tests (Playwright)

test('user can complete checkout', async ({ page }) => {
  await page.goto('/products');
  await page.getByRole('button', { name: 'Add to cart' }).first().click();
  await page.getByRole('link', { name: 'Cart' }).click();
  await expect(page.getByText('1 item')).toBeVisible();
  await page.getByRole('button', { name: 'Checkout' }).click();
});

React Query Patterns

function useUsers(filters: UserFilters) {
  return useQuery({
    queryKey: ['users', filters],
    queryFn: () => fetchUsers(filters),
    staleTime: 5 * 60 * 1000,
    placeholderData: keepPreviousData,
  });
}

function useUpdateUser() {
  const queryClient = useQueryClient();
  return useMutation({
    mutationFn: updateUser,
    onMutate: async (newUser) => {
      await queryClient.cancelQueries({ queryKey: ['users'] });
      const previous = queryClient.getQueryData(['users']);
      queryClient.setQueryData(['users'], (old) =>
        old.map(u => u.id === newUser.id ? { ...u, ...newUser } : u)
      );
      return { previous };
    },
    onError: (err, newUser, context) => {
      queryClient.setQueryData(['users'], context.previous);
    },
    onSettled: () => {
      queryClient.invalidateQueries({ queryKey: ['users'] });
    },
  });
}

Memoization Decision Table

TechniqueUse WhenDo NOT Use When
useMemoExpensive computation, referential equality for depsSimple calculations, primitive values
useCallbackFunctions passed to memoized childrenFunctions not passed as props
React.memoComponent re-renders often with same propsProps change on every render
NoneDefault — do not memoizeAlways profile first

Anti-Patterns / Common Mistakes

Anti-PatternWhy It Is WrongCorrect Approach
useEffect for data fetchingRace conditions, no caching, no dedupReact Query or Server Components
Prop drilling more than 2 levelsTight coupling, maintenance burdenComposition, context, or Zustand
Business logic in componentsUntestable, unreusableExtract to hooks or utility functions
Barrel exportsBreaks tree-shaking, slower buildsDirect imports
Testing implementation detailsBrittle tests that break on refactorTest behavior: user actions and outcomes
any type anywhereDefeats TypeScript's purposeunknown + type guards
Inline styles for non-dynamic valuesInconsistent, hard to maintainCSS modules, Tailwind, or styled-components
Memoizing everythingAdds complexity, often slowerProfile first, memoize second

Documentation Lookup (Context7)

Use mcp__context7__resolve-library-id then mcp__context7__query-docs for up-to-date docs. Returned docs override memorized knowledge.

  • react — when uncertain about hooks API, component lifecycle, or React 19+ features
  • next.js — for App Router, Server Components, or Next.js-specific APIs
  • typescript — for advanced type patterns or compiler options
  • tailwindcss — for utility classes, configuration, or plugin API
  • vitest — for test runner API, matchers, or mock utilities

Integration Points

SkillRelationship
testing-strategyStrategy defines frontend test frameworks
test-driven-developmentComponents are built with TDD cycle
react-best-practicesDetailed React patterns complement this skill
performance-optimizationFrontend performance follows optimization methodology
code-reviewReview verifies component architecture and test coverage
clean-codeCode quality principles apply to component code
webapp-testingPlaywright E2E tests use this skill's page structure
acceptance-testingUI acceptance criteria drive component tests

Key Principles

  • TypeScript strict mode, no any (use unknown + type guards)
  • Prefer composition over inheritance
  • Colocate tests, styles, and stories with components
  • Server Components by default; Client Components only when required
  • Error boundaries at route and feature boundaries
  • Accessibility is not optional (test with axe-core)

Skill Type

FLEXIBLE — Adapt component architecture and state management to the existing project conventions. The three-phase workflow is strongly recommended. Test coverage must target >85%. TypeScript strict mode is non-negotiable.

平台分布

Codex

33.93%
按下载量换算83

Claude

31.98%
按下载量换算79

Cursor

16.86%
按下载量换算41

Gemini CLI

9.57%
按下载量换算24

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills