nw-root-why

Code facts — resolve structural facts about code through code-fact CLI (vendor-neutral, bundled adapters: AST then TextSearch). Degrade LOUD. Never ad-hoc grep. Example: code-fact query.callers-of SYMBOL --root ROOT.

NW-ROOT-WHY: Toyota 5 Whys Root Cause Analysis

Wave: CROSS_WAVE Agent: Rex (nw-troubleshooter)

Overview

Systematic root cause analysis using Toyota's 5 Whys with multi-causal investigation and evidence-based validation. Investigates multiple cause branches at each level|validates solutions against all identified root causes.

Design-time counterexamples

RCA applies to design-time evidence at par with runtime failures: a model-checker violation, a refuted invariant, a failing property, or a theorem that only passes in conditional form. For these, the first RCA question is representational — which type or schema made the failing state expressible? A conditional theorem's hypothesis names the constraint the design should enforce by construction. Route the cure through nw-design's Counterexample discipline; a law added over an unchanged representation without recorded justification is a symptom patch, not a root-cause fix. This applies whether the counterexample came from a proof assistant, a model checker, or a property test in the project's own language — no prover is a prerequisite for this RCA question.

Agent Invocation

@nw-troubleshooter

Execute *investigate-root-cause for {problem-statement}.

Configuration:

• investigation_depth: 5
• multi_causal: true
• evidence_required: true

Usage: DELIVER Wave Retrospective (Phase 3.5)

When invoked as part of /nw-deliver Phase 3.5, execute in order:

1. Analyze What Worked Well — Identify practices that succeeded and document why — preserve these. Gate: at least one practice documented with rationale.
2. Analyze What Improved — Identify what worked better than before and why — record as reinforcements. Gate: delta from prior execution captured.
3. Analyze What Worked Badly — Apply 5 Whys; the root cause names the producer/type that admitted the failing state (GDP-0) — a fix that adds a check instead is a symptom patch unless that producer cannot change, with the reason recorded. Gate: each failure has a representational root cause and fix.
4. Analyze What Regressed — Apply 5 Whys to anything worse than before; produce prevention action. Gate: each regression has a root cause and prevention action.
5. Tag Meta-Improvements — Mark any items requiring nWave framework changes as meta-improvements. Gate: all framework-level issues flagged.
6. Append Retrospective — Write retrospective section to evolution document. If clean execution (no skips, no failures, no tooling issues), generate brief summary only. Gate: evolution document updated.