Token导航 LogoToken导航

Agent Skill · 研究检索

quality-assurance(品质保证)

quality-assurance 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

quality-assurance 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,权限为需要联网;许可证为 MIT,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 48,042。

总安装 706周安装 30GitHub Stars 636下载 247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
MIT

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:quality-assurance(品质保证)
来源仓库:https://github.com/jasonkneen/kiro
仓库路径:skills/quality-assurance
安装命令:
npx skills add https://github.com/jasonkneen/kiro --skill quality-assurance
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/jasonkneen/kiro --skill quality-assurance

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。

quality-assurance 用于查找、检索和筛选相关信息,适合在信息密集型任务中快速定位关键内容。适用于需要基于关键词或上下文筛选来源的场景,如文档调研或知识提取。通过指定查询条件调用检索接口,返回匹配结果列表。使用前请核实权限范围和维护状态,避免触发不必要的数据访问或命令执行。

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/quality-assurance

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/jasonkneen/kiro --skill quality-assurance;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

    它需要需要联网,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

quality-assurance 怎么安装?

在终端执行 npx skills add https://github.com/jasonkneen/kiro --skill quality-assurance。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

quality-assurance 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。

quality-assurance 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

quality-assurance 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

quality-assurance 用的人多吗?

公开数据是累计安装 706 次,近一周安装 30 次,下载 247 次,GitHub 636 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。

quality-assurance 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。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。
  • 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

  • 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。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 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
quality-assurance

品质保证 · 本页

706636需要联网MIT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

Quality Assurance

Ensure quality throughout the spec-driven development process with validation techniques, quality gates, and testing strategies.

When to Use This Skill

Use quality assurance practices when:

  • Completing any spec phase (requirements, design, tasks)
  • Transitioning between phases
  • Implementing features from specs
  • Reviewing completed work
  • Establishing team quality standards

Core Principles

  1. Requirements-Driven Testing: Every test traces to a requirement
  2. Phase-Appropriate Validation: Different techniques for each phase
  3. Continuous Quality: Checks throughout development
  4. Automated Where Possible: Reduce manual effort
  5. Fast Feedback: Catch issues early

Phase-Specific Validation

Requirements Phase Validation

Quality Checklist:

  • All user stories have acceptance criteria
  • Requirements are unambiguous and specific
  • Each requirement can be validated/tested
  • EARS format used consistently
  • Requirements link to business objectives
  • No conflicting requirements

Review Process:

  1. Self Review: Author checks completeness
  2. Stakeholder Review: Business validates requirements
  3. Technical Review: Team assesses feasibility
  4. Approval: Formal sign-off before design

Validation Techniques:

  • Scenario Walkthroughs: Step through user journeys
  • Edge Case Analysis: Identify boundary conditions
  • Conflict Detection: Check for contradictions
  • Completeness Analysis: Ensure all needs covered

Design Phase Validation

Quality Checklist:

  • Design addresses all requirements
  • Scalability considerations documented
  • Maintainability assessed
  • Security addressed
  • Performance requirements considered
  • External integrations defined

Review Process:

  1. Architecture Review: Senior team validates design
  2. Security Review: Security implications assessed
  3. Performance Review: Performance characteristics evaluated
  4. Integration Review: External dependencies validated

Validation Techniques:

  • Design Walkthroughs: Step through system interactions
  • Threat Modeling: Identify security vulnerabilities
  • Performance Modeling: Estimate system performance
  • Dependency Analysis: Map external requirements

Tasks Phase Validation

Quality Checklist:

  • Each task has clear deliverables
  • Task sequence is logical
  • All design elements covered
  • Each task can be validated
  • Tasks appropriately sized (2-4 hours)
  • Dependencies clearly defined

Review Process:

  1. Completeness Review: All design elements have tasks
  2. Sequencing Review: Task order is logical
  3. Scope Review: Tasks are appropriately sized
  4. Dependency Review: Dependencies clear

Quality Gates

Requirements Phase Exit Criteria

  • All user stories follow proper format
  • Acceptance criteria use EARS format
  • Requirements are testable and measurable
  • No conflicting requirements
  • Stakeholders have approved
  • Edge cases documented

Design Phase Exit Criteria

  • Architecture addresses all requirements
  • Non-functional requirements addressed
  • External dependencies identified
  • Data models clearly defined
  • Error handling documented
  • Security considerations addressed
  • Technical review completed

Tasks Phase Exit Criteria

  • All design elements have tasks
  • Tasks properly sequenced
  • Each task is actionable
  • Requirements references included
  • Test approach defined
  • Task breakdown reviewed

Task-Level Quality Gates

Before Starting:

  • Task requirements understood
  • Test strategy defined
  • Dependencies available
  • Environment ready

During Implementation:

  • Code follows standards
  • Tests written alongside code
  • Coverage meets thresholds (80%+)
  • No critical vulnerabilities

Before Marking Complete:

  • All tests pass
  • Code review completed
  • Documentation updated
  • Requirements validated

Testing Strategies

Test Pyramid

       /\
      /  \     E2E Tests (10%)
     /____\    Integration Tests (20%)
    /      \
   /________\   Unit Tests (70%)

Unit Testing

  • Fast execution (< 1 second per test)
  • Test individual functions/classes
  • Mock external dependencies
  • Target 80%+ coverage

Integration Testing

  • Test component interactions
  • Use real dependencies where practical
  • Validate API contracts
  • Test critical workflows

End-to-End Testing

  • Test complete user journeys
  • Production-like environment
  • Focus on critical business flows
  • Minimal but comprehensive

Test-Driven Development

For each task:

  1. Write tests first based on acceptance criteria
  2. Run tests - verify they fail (red)
  3. Write code - minimal to pass tests (green)
  4. Refactor - improve while keeping tests green
  5. Validate - ensure requirements satisfied

Quality Metrics

Code Quality

  • Line Coverage: % of code lines executed
  • Branch Coverage: % of code branches tested
  • Cyclomatic Complexity: Code complexity
  • Technical Debt: Accumulated issues

Testing Effectiveness

  • Test Pass Rate: % of tests passing
  • Execution Time: Time to run tests
  • Defect Detection Rate: Bugs found by tests vs production
  • Test Maintenance: Time spent maintaining tests

Common Quality Issues

Flaky Tests

Symptoms: Tests pass/fail inconsistently Solutions:

  • Identify timing dependencies
  • Use proper wait conditions
  • Isolate test data
  • Fix race conditions

Slow Test Suites

Symptoms: Tests take too long Solutions:

  • Parallelize execution
  • Optimize database operations
  • Use test doubles for external services
  • Profile and optimize slow tests

Low Coverage

Symptoms: Insufficient code coverage Solutions:

  • Add tests for uncovered paths
  • Focus on critical business logic
  • Use mutation testing
  • Set coverage gates in CI

Test Maintenance Burden

Symptoms: Tests require frequent updates Solutions:

  • Improve test design
  • Use page object patterns for UI
  • Reduce coupling to implementation
  • Regular test refactoring

Validation Checklists

Requirements Validation

## Requirements Review

**Completeness:**
- [ ] All user roles addressed
- [ ] Happy path scenarios covered
- [ ] Edge cases documented
- [ ] Error cases handled
- [ ] Business rules captured

**Clarity:**
- [ ] Precise language used
- [ ] No ambiguous terms
- [ ] Technical jargon avoided
- [ ] Behaviors are specific

**Testability:**
- [ ] Each requirement verifiable
- [ ] Success criteria observable
- [ ] Inputs/outputs specified

Design Validation

## Design Review

**Architecture:**
- [ ] Requirements addressed
- [ ] Components well-defined
- [ ] Interfaces specified
- [ ] Data flow documented

**Quality Attributes:**
- [ ] Performance considered
- [ ] Security addressed
- [ ] Scalability planned
- [ ] Maintainability assessed

**Risks:**
- [ ] Single points of failure identified
- [ ] Bottlenecks documented
- [ ] Mitigations planned

Implementation Validation

## Implementation Review

**Code Quality:**
- [ ] Follows standards
- [ ] Well-documented
- [ ] Tests included
- [ ] No security issues

**Requirements:**
- [ ] All criteria met
- [ ] Edge cases handled
- [ ] Error handling complete

**Integration:**
- [ ] Works with existing code
- [ ] APIs functioning
- [ ] Performance acceptable

Best Practices

Testing Best Practices

  1. Write tests first when possible
  2. Each test verifies one thing
  3. Use descriptive test names
  4. Maintain test independence
  5. Keep tests up-to-date

Quality Assurance Best Practices

  1. Find issues early (shift left)
  2. Automate everything possible
  3. Use metrics to drive improvements
  4. Make quality everyone's responsibility
  5. Continuous learning and improvement

Process Integration

  1. Link tests to requirements
  2. Provide quick feedback
  3. Focus testing on high-risk areas
  4. Keep documentation current
  5. Integrate tools with workflow

Quick Reference

Phase Transition Questions:

  • Requirements → Design: "Does design address all requirements?"
  • Design → Tasks: "Do tasks cover all design elements?"
  • Tasks → Implementation: "Does code satisfy task specifications?"

Quality Gate Questions:

  • "Is this testable and measurable?"
  • "Have we considered what could go wrong?"
  • "Would another developer understand this?"
  • "Does this meet our standards?"

平台分布

OpenCode

30.4%
按下载量换算75

windsurf

20.65%
按下载量换算51

Claude Code

15.63%
按下载量换算39

Codex

11.31%
按下载量换算28

Gemini CLI

6.95%
按下载量换算17

Antigravity

3.06%
按下载量换算8

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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。

继续浏览同类 Skills