Token导航 LogoToken导航TokenDH.com
研究检索操作浏览器github未标认证来源可访问许可证需确认审计通过

test-review测试回顾

Agent Skill

用于辅助测试设计、自动化测试、用例整理和回归验证。它适合让 Agent 编写单元测试、端到端测试、测试计划或根据失败日志定位问题。使用时需要确认项目测试框架、运行命令和夹具数据,避免为了通过测试而改坏真实逻辑;涉及浏览器或外部服务时,应区分本地模拟、测试环境和生产环境。

总安装

282

周安装

12

GitHub Stars

公开资料未说明

下载量

99
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

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:test-review(测试回顾)
来源仓库:https://github.com/cerico/macfair
仓库路径:skills/test-review
安装命令:
npx skills add https://github.com/cerico/macfair --skill test-review
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/cerico/macfair --skill test-review

简介

Test Review 审查功能相关测试,发现覆盖率缺口和质量问题。

  • 适用于 PR 前验证测试完整性,识别缺失的边界条件或无效断言。
  • 检查验证模式、边缘案例和错误处理路径是否充分覆盖。
  • 安装前需确认项目使用的测试框架(vitest/playwright),否则无法准确评估。
  • 输出为具体改进建议,需人工判断是否采纳并调整测试用例。

SKILL.md

Test Review

Review tests for a feature to catch coverage gaps and quality issues before PR.

Instructions

  1. Identify the feature being tested (from user or recent git changes)
  2. Find related source files: validations, tRPC routers, components, pages
  3. Find related test files: vitest and playwright
  4. Check coverage gaps - what's missing?
  5. Check test quality - are assertions meaningful?
  6. Check for common mistakes
  7. Report findings with file:line references

Coverage Gaps

Validation Schemas

  • Happy path with valid input
  • Each required field missing
  • Invalid formats (email, date, etc.)
  • Edge cases (empty string, zero, negative numbers)
  • Type coercion if applicable

tRPC Procedures

  • Success case returns expected shape
  • Error case (not found, unauthorized)
  • Input validation rejects bad data
  • Side effects happen (database updated, email sent)

React Components

  • Renders without crashing
  • Displays data correctly
  • Loading state
  • Error state
  • User interactions (click, submit, type)
  • Accessibility (keyboard nav, screen reader)

E2E (Playwright)

  • Happy path flow start to finish
  • Form validation errors shown to user
  • Error recovery (retry, go back)
  • Empty states

Test Quality

Meaningful Assertions

// Bad - just checks truthy
expect(result).toBeTruthy()

// Good - checks specific value
expect(result.id).toBe('123')
expect(result.items).toHaveLength(3)

Independent Tests

  • No shared mutable state between tests
  • Each test sets up its own data
  • Order of tests doesn't matter

Realistic Mocks

// Bad - mock returns nothing useful
const mockFetch = jest.fn()

// Good - mock returns realistic data
const mockFetch = jest.fn().mockResolvedValue({
  id: '123',
  name: 'Test User',
  createdAt: new Date(),
})

Error Testing

// Bad - just checks it throws
expect(() => fn()).toThrow()

// Good - checks specific error
expect(() => fn()).toThrow('User not found')
expect(() => fn()).toThrow(NotFoundError)

Common Mistakes

General

  • No .only or .skip left in code
  • No console.log in tests
  • No hardcoded IDs that won't exist in CI
  • No time-dependent tests without mocking Date.now()

Async Issues

// Bad - no await, test passes before async completes
it('creates user', () => {
  createUser({ name: 'Test' })
  expect(db.users).toHaveLength(1)
})

// Good - awaits async operation
it('creates user', async () => {
  await createUser({ name: 'Test' })
  expect(db.users).toHaveLength(1)
})

Playwright-Specific

  • Uses semantic locators (getByRole, getByLabel) not CSS selectors
  • No page.waitForTimeout() - use waitForResponse, waitForSelector, or expect().toBeVisible()
  • Tests don't depend on previous test's state
  • Assertions check visible outcomes, not implementation
// Bad - fragile selector, arbitrary timeout
await page.click('.btn-submit')
await page.waitForTimeout(1000)

// Good - semantic locator, waits for outcome
await page.getByRole('button', { name: 'Submit' }).click()
await expect(page.getByText('Saved successfully')).toBeVisible()

Vitest-Specific

  • Mocks are reset between tests (vi.clearAllMocks() in beforeEach)
  • No network calls without MSW or manual mocking
  • describe blocks group related tests logically

Output Format

## Test Review: [feature]

### Coverage Gaps
- validations/booking.ts has no tests for `updateBookingSchema`
- No E2E test for delete flow
- Missing error state test for BookingForm component

### Quality Issues
- tests/vitest/booking.test.ts:23 - assertion just checks truthy, verify specific value
- tests/vitest/booking.test.ts:45 - mock returns empty object, use realistic data

### Common Mistakes
- tests/playwright/booking.spec.ts:12 - uses `waitForTimeout(2000)`, use `waitForResponse` or assertion
- tests/vitest/booking.test.ts:67 - `.only` left in code

### Passed
- Validation schema has happy path + required field tests
- tRPC procedures test success and not-found cases
- No hardcoded IDs
- Semantic locators used throughout

Notes

  • Focus on changed files first, but check related test files too
  • Not every gap needs fixing - use judgment on risk vs effort
  • Flag .only and .skip as blockers - these break CI

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

33.29%
按下载量换算33

Claude

30.29%
按下载量换算30

Cursor

20.66%
按下载量换算20

Gemini CLI

10.12%
按下载量换算10

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills