RootCauseAnalysis

How This Skill Approaches It

Five workflows cover different failure shapes. FiveWhys builds a linear or branching causal chain, forcing you past the proximate cause until you reach something you can actually fix. Fishbone maps the 6 M's (Manpower, Machine, Method, Material, Measurement, Mother-Nature) or 4 P's when multiple cause categories are suspected. Postmortem structures a blameless timeline with contributing factors and action items — the format teams can actually be honest in. FaultTree uses AND/OR gate logic for safety-critical failures where multiple independent paths could each trigger the outcome. KepnerTregoe IS/IS-NOT builds a distinction table for the subtle defect that only appears in CI, only on Tuesdays, only with parallel workers — the kind of bug that defeats guesswork. For non-trivial incidents, Postmortem wraps the others: start with the blameless timeline, pull in FiveWhys or Fishbone as investigation tools inside it. The skill operates on five axioms: proximate cause is where analysis starts; incidents have multiple contributing factors, not one; humans are never root causes; you stop at a cause you can act on; and RCA is a bias-fight.

In Action

What you say to your DA, and what the RootCauseAnalysis skill actually does.

• You say "the payments service went down for 14 minutes last night, help me write the postmortem"
  → Runs Postmortem workflow: reconstructs the blameless timeline from deploy to rollback, runs FiveWhys inside to trace the latency spike to its contributing factors, identifies both active failures and latent conditions, and produces action items that prevent the failure class — not just the specific incident.
• You say "we've fixed this auth bug three times and it keeps coming back, what's actually wrong"
  → Runs Fishbone workflow: expands across all 6 M categories to find why no single fix has held, surfaces the latent conditions (likely Method + Measurement) that let the bug regenerate, and identifies the structural change that would actually close the loop.
• You say "this test only fails in CI, never locally — what's the actual cause"
  → Runs KepnerTregoe workflow: builds an IS/IS-NOT distinction table across environment, timing, concurrency, and filesystem axes to isolate the specific combination triggering the failure — rather than guessing at environment differences.

Inside the Skill

The thinking, frameworks, and architecture that distinguish this skill from a generic version of the same task.

User: "Payments threw 500s for ten minutes after last night's deploy. Write the postmortem."
→ Postmortem intent, with FiveWhys or Fishbone on each contributing thread
→ A blameless record with a UTC timeline, defensive-layer map and strength-rated owned actions

User: "The export job fails only on the EU cluster and only on Mondays. Why?"
→ KepnerTregoe intent
→ An IS/IS-NOT matrix, the change that explains every IS without predicting the IS-NOTs, and a test to confirm it

User: "5 whys on why the nightly backup keeps missing its window"
→ FiveWhys intent
→ An evidenced chain to systemic causes, read forward with no gaps, and owned corrective actions

Workflows & Routing · 5

Each workflow is one job the skill runs. The trigger phrases route your request to the right one — this is the skill's routing table.

1. 01
   FiveWhys Intents/FiveWhys.md

   5 whys, five whys, quick causal chain, ask why until root

2. 02
   Fishbone Intents/Fishbone.md

   fishbone, ishikawa, categorized cause map, 6 Ms / 4 Ps / 8 Ms

3. 03
   Postmortem Intents/Postmortem.md

   postmortem, incident review, blameless postmortem, production incident

4. 04
   FaultTree Intents/FaultTree.md

   fault tree, fta, top-down deductive, safety-critical, AND/OR logic

5. 05
   KepnerTregoe Intents/KepnerTregoe.md

   kepner tregoe, is/is-not, what changed, distinction analysis, subtle defects

How to Invoke

Say any of these to your DA and LifeOS activates the RootCauseAnalysis skill automatically:

• "root cause"
• "RCA"
• "5 whys"
• "fishbone"
• "postmortem"
• "incident analysis"
• "fault tree"
• "why does this keep failing"
• "blameless"
• "recurring bug"

Or invoke explicitly:

Skill("RootCauseAnalysis")

References · 2

Auxiliary files the skill loads at runtime — frameworks, guides, configs.

• Foundation
• MethodSelection

Related Skills