메뉴
Use when designing, checking, or interpreting A/B tests and experiments. Reviews hypothesis, variants, metrics, sample ratio, significance, guardrail metrics, segmentation, validity risks, and business recommendations.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Use when writing a concise daily report, daily worklog, standup update, shift handoff, or end-of-day summary for data analysis, data development, BI, governance, or operations work.
Use when writing or improving data documentation for warehouse tables, metrics, dashboards, SQL jobs, data products, or handover materials so other people and agents can understand data purpose, grain, fields, logic, ownership, and usage constraints.
Use when analyzing upstream and downstream impact for table changes, metric changes, field deprecation, pipeline migration, dashboard breakage, data governance review, or data asset dependency communication.
Use when planning integration with databases, BI tools, data platforms, APIs, workflow systems, storage services, or third-party tools, including connection requirements, API availability, auth strategy, configuration questions, and validation steps.
Use when planning or writing market research for a specific industry, product category, competitor set, business opportunity, user segment, or go-to-market question with structured evidence and clear assumptions.
Use when diagnosing business metric changes or anomalies such as GMV decline, conversion drop, retention loss, churn increase, revenue change, lead decline, or active user fluctuation with a structured root cause analysis plan.
| name | ab-test-analysis |
| description | Use when designing, checking, or interpreting A/B tests and experiments. Reviews hypothesis, variants, metrics, sample ratio, significance, guardrail metrics, segmentation, validity risks, and business recommendations. |
设计、审查或解读 A/B 实验，判断实验是否可信、结果是否显著、业务上是否值得推广。
这个 Skill 关注实验分析方法，不负责替代产品决策。实验结论必须同时看统计结果、业务影响和实验有效性。
使用这个 Skill，当用户需要：
典型输入：
这个实验转化率提升了 0.3 个百分点，p 值 0.04，可以全量吗？
不要用这个 Skill 直接完成：
funnel-analysisretention-cohort-analysismetric-definition-reviewerdata-analysis-report-writer如果没有实验组、对照组和核心指标，不要强行给实验结论。
最少输入：
可选输入：
推荐使用 实验分析上下文模板。
最有价值的上下文是：实验假设、分组方式、核心指标、护栏指标、样本量、实验周期和显著性结果。缺少护栏指标或分流说明时，不能直接建议全量。
## A/B 实验分析
### 1. 实验结论
### 2. 实验设计检查
| 项目 | 结果 | 风险 |
| --- | --- | --- |
| 实验假设 | | |
| 分组方式 | | |
| 实验周期 | | |
| 样本量 | | |
| 核心指标 | | |
| 护栏指标 | | |
### 3. 指标结果
| 指标 | 对照组 | 实验组 | 变化 | 显著性 | 解释 |
| --- | --- | --- | --- | --- | --- |
### 4. 有效性风险
### 5. 分群结果
### 6. 决策建议
### 7. 需要补充的数据
输出必须：
请用 ab-test-analysis 解读这个实验：
假设：新版会员页能提高开通转化率
对照组：旧版页面，10000 UV，转化 310
实验组：新版页面，10050 UV，转化 355
护栏指标：退款率、页面加载时间、客服投诉
周期：7 天