RootCauseAnalysis

Investigate failures using Five Whys, Fishbone, Postmortem, Fault Tree, and Kepner-Tregoe methods.

Updated Sep 3, 2026
One-click install
npx skills add https://github.com/ruban-s/DevOS --skill rootcauseanalysis-ruban-s
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: RootCauseAnalysis
Source: https://github.com/ruban-s/DevOS/tree/main/skills/RootCauseAnalysis
Command: npx skills add https://github.com/ruban-s/DevOS --skill rootcauseanalysis-ruban-s

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When incidents, outages, or recurring bugs strike, teams often stop at the first convenient explanation or blame a person, letting the same failure family return. This Skill structures incident investigation so inquiries trace failures past the trigger to systemic, fixable root causes. ## Core Features & Use Cases - Five investigation workflows: Five Whys for single-thread breaks, Fishbone for multi-family cause mapping with Pareto ordering, blameless Postmortem for formal incident records, Fault Tree Analysis for safety-critical Boolean decomposition, and Kepner-Tregoe IS/IS-NOT for elusive environment-specific defects. - Method selection guidance: Routing tables and stacking patterns match the right technique to each break shape, clock budget, and safety weight. - Bias and blame defenses: Built-in guards against hindsight bias, single-cause framing, and human-blame halting, plus a remedy strength ladder favoring automation over reminders. - Use Case: After a payments outage, run the Postmortem workflow to build a forward timeline, apply 5 Whys to each thread, chart layered-defense holes, and produce a blameless report with owned, dated follow-up actions. ## Quick Start Ask the agent to run a root cause analysis on a recent incident, for example by saying the payments service went down for 14 minutes last night and you need a blameless postmortem with contributing causes and fixes.

Frequently Asked Questions about RootCauseAnalysis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a root cause analysis on a production incident?▼

Start with the Postmortem workflow: rebuild the timeline forward before theorizing, then run 5 Whys or Fishbone inside it per thread. Record contributors in plural, chart layered-defense holes, and assign owned, dated follow-up actions ordered by remedy strength.

What is the difference between 5 Whys and Fishbone analysis?▼

5 Whys walks one causal thread deep, suiting single-thread breaks with a known trigger. Fishbone sweeps wide first, sorting causes into families like the 6 M's, then uses Pareto to prioritize before drilling into the vital few.

When should I use Kepner-Tregoe instead of 5 Whys?▼

Use Kepner-Tregoe IS/IS-NOT for shy, intermittent, or environment-specific defects where something works in one context but not another. It requires a known-good baseline to contrast against; without one, use Fishbone or Apollo-style charting instead.

When is Fault Tree Analysis the right choice?▼

Fault Tree Analysis fits safety-critical or security-critical systems with redundant defenses and multi-path failures, especially when probability estimates are needed. It is time-intensive (hours to days), so for fast triage use 5 Whys instead.

Why does root cause analysis keep ending at human error?▼

Halting at a person is a known failure mode: a human slip always stands on a setup that permitted or invited it. The method requires asking one more why to expose the systemic condition, keeping the inquiry blameless and fixable.

What are the limitations of the 5 Whys method?▼

5 Whys cannot handle branching causes, cross-service failures, or probability estimation, and it cannot outrun the investigator's domain knowledge. For tangled distributed failures, use evidence-per-node charting or Fault Tree Analysis instead.