Token导航 LogoToken导航TokenDH.com
研究检索需要联网github未标认证来源可访问许可证需确认审计通过

five-whys五个为什么

Agent Skill

five-whys 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。

总安装

840

周安装

34

GitHub Stars

235

下载量

264
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:five-whys(五个为什么)
来源仓库:https://github.com/flpbalada/my-opencode-config
仓库路径:skills/five-whys
安装命令:
npx skills add https://github.com/flpbalada/my-opencode-config --skill five-whys
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/flpbalada/my-opencode-config --skill five-whys

简介

five-whys 用于查找、检索和筛选相关信息。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词快速定位候选结果。
  • 通过 npx skills add 命令从指定仓库安装,需确认权限与维护状态。
  • 使用前应核实是否会触发联网、命令执行或文件读写操作。
  • 建议结合原始 README 进一步核验具体用法和功能边界。

SKILL.md

Five Whys - Root Cause Analysis

Systematic guide to uncovering root causes through iterative questioning, originally developed by Sakichi Toyoda for Toyota Motor Corporation.

When to Use This Skill

  • Investigating recurring problems
  • Debugging system failures
  • Understanding customer churn
  • Analyzing project delays or budget overruns
  • Post-mortem analysis
  • Process improvement initiatives

Core Concepts

The Method

Problem Statement
    ↓
Why? → Answer 1
    ↓
Why? → Answer 2
    ↓
Why? → Answer 3
    ↓
Why? → Answer 4
    ↓
Why? → Answer 5
    ↓
Root Cause Identified
    ↓
Solution Implementation

Key Principles

PrincipleDescription
Facts over assumptionsBase questions on data, not guesses
Systems over individualsFocus on process failures, not blame
FlexibilityGo beyond 5 questions if needed
VerificationValidate findings with evidence

Questioning Techniques

Standard Approach

For each iteration, ask:

  • Why did this happen?
  • What caused this situation?
  • What led to this outcome?

Alternative Phrasing (Less Confrontational)

When direct "why" questions feel threatening, use softer alternatives:

Root Cause Investigation

Instead of...Try...
Why did this happen?What was going on when this happened?
Why did you do that?What were you trying to accomplish?
Why is this broken?How do you suppose we ended up here?

Understanding Motivation

Instead of...Try...
Why do you want this?What happens if we don't get this done?
Why does this matter?What problems does this solve?
Why is this urgent?What do you think will happen if we delay?

Understanding Decisions

Instead of...Try...
Why did leadership decide this?What were the reasons we went this direction?
Why this approach?How do you see this working long term?

Analysis Framework

Step 1: Define the Problem

Be specific and measurable:

❌ Bad:  "The system is slow"
✅ Good: "Page load time increased from 2s to 8s after the March release"

❌ Bad:  "Customers are unhappy"
✅ Good: "Customer churn increased by 40% over three months"

Step 2: Iterate Through Whys

Document each level clearly:

Problem: Customer churn increased by 40%

1. Why? → Customers canceling after free trial
2. Why? → Not seeing enough value during trial
3. Why? → Not completing the onboarding process
4. Why? → Onboarding too complex, requires too much setup
5. Why? → Product lacks automation and intelligent defaults

Root Cause: Poor onboarding experience due to lack of automation

Step 3: Identify Solutions

Target the root cause, not symptoms:

Symptom-level fix (avoid):
├── Offer discounts to retain customers
└── Send more reminder emails

Root cause fix (preferred):
├── Build automated data import
├── Create intelligent defaults by industry
├── Simplify onboarding to 3 steps
└── Add progress indicators

Output Template

After completing analysis, document as:

## Five Whys Analysis

**Problem Statement:** [Clear, measurable problem description]

**Analysis Date:** [Date]

**Participants:** [Who was involved]

### Question Chain

1. **Why?** [First answer with evidence]
2. **Why?** [Second answer with evidence]
3. **Why?** [Third answer with evidence]
4. **Why?** [Fourth answer with evidence]
5. **Why?** [Fifth answer with evidence]

### Root Cause

[Identified root cause - the systemic issue to address]

### Recommended Solutions

| Priority | Solution     | Expected Impact | Effort   |
| -------- | ------------ | --------------- | -------- |
| High     | [Solution 1] | [Impact]        | [Effort] |
| Medium   | [Solution 2] | [Impact]        | [Effort] |
| Low      | [Solution 3] | [Impact]        | [Effort] |

### Success Metrics

- [How will we measure if the solution worked?]

Classic Examples

Manufacturing Example (Toyota Original)

Problem: Machine stopped operating

1. Why? → Motor overheated
2. Why? → Wasn't lubricated enough
3. Why? → Oil pump failed
4. Why? → Filter was clogged
5. Why? → No regular maintenance schedule

Root Cause: Lack of preventive maintenance procedures
Solution: Implement maintenance schedule and checklist

SaaS Example

Problem: Customer churn increased 40%

1. Why? → Customers canceling after free trial
2. Why? → Not seeing enough value during trial
3. Why? → Not completing onboarding
4. Why? → Onboarding too complex
5. Why? → Lacks automation and smart defaults

Root Cause: Poor onboarding experience
Solutions:
- Automated data import from popular tools
- Intelligent defaults based on industry
- Simplified 3-step onboarding
- In-app progress indicators

Result: 60% decrease in churn, 35% increase in trial conversion

Software Bug Example

Problem: Production API returning 500 errors

1. Why? → Database queries timing out
2. Why? → Query taking 30+ seconds
3. Why? → Missing index on frequently queried column
4. Why? → Index was dropped during migration
5. Why? → Migration script lacked index recreation step

Root Cause: Incomplete migration testing process
Solutions:
- Add index verification to migration checklist
- Implement automated index coverage tests
- Create pre-production performance benchmarks

Best Practices

Do

  • Use data - Support answers with evidence and metrics
  • Involve diverse perspectives - Different viewpoints reduce blind spots
  • Focus on systems - Ask "what process failed?" not "who failed?"
  • Document everything - Create audit trail for future reference
  • Verify root cause - Test that fixing it would prevent recurrence

Avoid

  • Stopping too early - Surface answers are usually symptoms
  • Personal blame - "John made a mistake" is never the root cause
  • Single path - Complex problems may have multiple root causes
  • Assumptions - Always verify with data
  • Skipping steps - Each "why" should logically follow the previous

Communication Tips

  • Give people time to respond - embrace silence
  • Ask one question at a time
  • Resist the urge to clarify before they answer
  • Focus on curiosity rather than interrogation
  • Frame as collaborative problem-solving

Integration with Other Methods

The Five Whys works well alongside:

MethodCombined Use
KaizenContinuous improvement cycles
Six SigmaDMAIC problem-solving
Fishbone DiagramVisualizing multiple cause categories
Pareto AnalysisPrioritizing which problems to analyze
Post-mortemIncident review sessions

Resources

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

34.26%
按下载量换算90

Claude

31.42%
按下载量换算83

Cursor

16.82%
按下载量换算44

Gemini CLI

8.89%
按下载量换算23

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。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