一键在 Manus 中运行任何 Skill

root-cause-investigation

星标3

分支3

更新时间2026年6月14日 05:12

Systematically diagnose why a metric changed unexpectedly. Activate when a KPI moves significantly, a stakeholder asks "why did X drop/spike?", or a post-incident review requires evidence-based root cause documentation.

安装

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

在 Manus 中运行

来源

ryudkiss-hue

ryudkiss-hue/nyc_data

打开 GitHub 仓库查看创作者相关仓库

下载

在 Manus 中运行

When to use

A key metric moved significantly and the team needs an explanation
A stakeholder challenges a number and you need to trace the cause
Post-incident review requires documented root cause analysis
Distinguishing real signal from noise in a reported metric change

Process

Validate the change — confirm the metric movement exceeds normal variance (use rolling average ± 2σ); rule out data quality issues before proceeding
Establish timing — plot the full metric history; identify exactly when the change began; distinguish sudden shift from gradual trend
Decompose the metric — break the metric into its components (e.g. revenue = volume × price × mix); use scripts/drilldown_analyzer.py to identify which component(s) drove the change
Drill down by dimension — compare performance across geography, platform, user segment, product; rank contributors by impact (absolute + % of total change)
Generate and test hypotheses — correlate the change timing against a list of known events (releases, campaigns, incidents); use references/hypothesis_testing_guide.md
Document findings — fill assets/rca_report_template.md with: change magnitude, timing, component analysis, dimension drilldown, validated hypothesis, and tiered recommendations

Inputs the skill needs

Required: metric name, time period, and magnitude of change
Required: historical data sufficient to establish a baseline
Optional: dimensional breakdown data (by segment, geography, channel)
Optional: event log of potential causal events (releases, campaigns, outages)

Output

scripts/drilldown_analyzer.py — metric decomposition and dimensional contribution analysis
references/rca_framework.md — structured RCA methodology and decision tree
references/hypothesis_testing_guide.md — common root cause patterns and diagnostic tests
assets/rca_report_template.md (filled) — evidence-based findings with tiered recommendations

同仓库更多 Skills

同仓库

data-catalog-entry

ryudkiss-hue/nyc_data

Create standardized metadata for data assets. Use when documenting new datasets, building data catalogs, improving data discoverability, or creating data dictionaries for teams.

2026-06-173

analysis-planning

ryudkiss-hue/nyc_data

Structure analysis approach before starting work. Use when receiving new analysis requests, breaking down complex questions into steps, or planning iterative analysis workflows.

2026-06-173

data-quality-audit

ryudkiss-hue/nyc_data

Comprehensive data quality assessment against business rules, schema constraints, and freshness expectations. Activate when validating production pipelines, onboarding new data sources, or when stakeholders report data quality concerns.

2026-06-143

segmentation-analysis

ryudkiss-hue/nyc_data

Identify distinct customer or user segments based on behavior, attributes, or value. Activate when you need to answer "who are our best customers?" or "what distinct groups exist in our user base?" and need data-driven profiles to inform strategy.

2026-06-143

visualization-builder

ryudkiss-hue/nyc_data

Create effective, publication-ready data visualizations. Use when choosing chart types, designing presentation visuals, building dashboard charts, or applying visual design best practices to data output.

2026-06-143

methodology-explainer

ryudkiss-hue/nyc_data

Create a clear, transparent explanation of analytical methodology for any audience level. Activate when you deliver findings that require the audience to trust the method — A/B tests, attribution models, forecasts, statistical analyses, or anything where "how did you get that?" is a likely question.

2026-06-143

name	root-cause-investigation
description	Systematically diagnose why a metric changed unexpectedly. Activate when a KPI moves significantly, a stakeholder asks "why did X drop/spike?", or a post-incident review requires evidence-based root cause documentation.

When to use

A key metric moved significantly and the team needs an explanation
A stakeholder challenges a number and you need to trace the cause
Post-incident review requires documented root cause analysis
Distinguishing real signal from noise in a reported metric change

Process

Validate the change — confirm the metric movement exceeds normal variance (use rolling average ± 2σ); rule out data quality issues before proceeding
Establish timing — plot the full metric history; identify exactly when the change began; distinguish sudden shift from gradual trend
Decompose the metric — break the metric into its components (e.g. revenue = volume × price × mix); use scripts/drilldown_analyzer.py to identify which component(s) drove the change
Drill down by dimension — compare performance across geography, platform, user segment, product; rank contributors by impact (absolute + % of total change)
Generate and test hypotheses — correlate the change timing against a list of known events (releases, campaigns, incidents); use references/hypothesis_testing_guide.md
Document findings — fill assets/rca_report_template.md with: change magnitude, timing, component analysis, dimension drilldown, validated hypothesis, and tiered recommendations

Inputs the skill needs

Required: metric name, time period, and magnitude of change
Required: historical data sufficient to establish a baseline
Optional: dimensional breakdown data (by segment, geography, channel)
Optional: event log of potential causal events (releases, campaigns, outages)

Output

scripts/drilldown_analyzer.py — metric decomposition and dimensional contribution analysis
references/rca_framework.md — structured RCA methodology and decision tree
references/hypothesis_testing_guide.md — common root cause patterns and diagnostic tests
assets/rca_report_template.md (filled) — evidence-based findings with tiered recommendations