RootCauseAnalysis

Convert incident reports into blameless root-cause analyses with RCA workflows.

1|Updated Apr 16, 2026
One-click install
npx skills add https://github.com/davdunc/pai-framework --skill rootcauseanalysis-davdunc
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: RootCauseAnalysis
Source: https://github.com/davdunc/pai-framework/tree/main/skills/RootCauseAnalysis
Command: npx skills add https://github.com/davdunc/pai-framework --skill rootcauseanalysis-davdunc

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill turns incident symptoms and “proximate causes” into a structured, blameless analysis that identifies actionable contributing factors and latent system defenses.

Core Features & Use Cases

  • Wrap incident learning as a blameless postmortem: reconstruct timelines, classify proximate vs contributing factors, and produce strong action items with owners and verification.
  • Choose the right RCA workflow: run FiveWhys for causal chains, Fishbone for category mapping, Kepner-Tregoe for IS/IS-NOT deviations, FaultTree for safety-critical Boolean/AND-OR structure, and Postmortem as the orchestrating wrapper.
  • Validate and prioritize: enforce an “actionability test,” reduce bias (hindsight/confirmation/single-cause), and prioritize remediation based on where a change can prevent recurrence.

Quick Start

Use the RootCauseAnalysis skill to generate a blameless postmortem with contributing factors and specific action items by running it on your incident description.

Frequently Asked Questions about RootCauseAnalysis

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

FAQPage Schema
How do I run a blameless postmortem for a production outage?

A blameless postmortem reconstructs incident timelines and classifies proximate vs contributing factors to generate actionable fixes. It enforces an actionability test to prioritize systemic interventions preventing recurrence without assigning individual blame.

What is the difference between Five Whys and a Fishbone diagram for incident investigation?

Five Whys traces linear causal chains, while a Fishbone diagram maps contributing factors across operational categories. Both root cause analysis workflows feed into a broader postmortem to validate bias-resistant, actionable systemic remedies.

When should I use Fault Tree analysis instead of Kepner-Tregoe for a security incident?

Use Fault Tree analysis for safety-critical Boolean AND-OR logic structures in security incidents, and Kepner-Tregoe for IS/IS-NOT deviation mapping. Fault Tree isolates critical system defenses, while Kepner-Tregoe narrows down specific incident deviations.

Can I analyze recurring software defects using IS/IS-NOT analysis?

Yes, IS/IS-NOT analysis maps deviations to distinguish what the defect is from what it is not. This Kepner-Tregoe workflow reduces confirmation bias and isolates specific recurring software defect conditions to produce targeted action items.

How do I write action items for incident investigation that prevent recurrence?

Write action items by applying an actionability test to validated contributing factors, ensuring each item has a specific owner and verification method. This prioritizes systemic, implementable interventions targeting latent system defenses where changes prevent recurrence.

Does root cause analysis help with pre-launch risk inversion for operational systems?

Yes, root cause analysis applies to pre-launch risk inversion by identifying latent system defenses and contributing factors before deployment. It converts potential operational system failures into prioritized, actionable interventions using postmortem workflows.