but-why

Guide structured, evidence-driven questioning sessions to uncover root causes.

6|Updated Mar 18, 2026
One-click install
npx skills add https://github.com/anotherben/claude-harness --skill but-why
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: but-why
Source: https://github.com/anotherben/claude-harness/tree/main/skills/but-why
Command: npx skills add https://github.com/anotherben/claude-harness --skill but-why

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill helps teams uncover root causes and avoid surface-level explanations by guiding structured, evidence-driven questioning.

Core Features & Use Cases

  • Seed the objective and keep questions traceable to that goal.
  • The questioning loop focuses on one question at a time, progressively revealing deeper causes.
  • Branching strategy and evidence gathering help manage multiple threads and verify assumptions.
  • Contradiction detection and bedrock-focused summaries ensure accountability and learning.

Quick Start

Kick off a but-why session by defining the objective and then asking one focused question to begin the investigation.

Frequently Asked Questions about but-why

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

FAQPage Schema
How do I find the root cause of a software bug when surface explanations are insufficient?

To find the root cause of a software bug, you use structured, evidence-driven questioning that uncovers deeper causes. This method enforces an objective seed and a one-question-at-a-time loop to progressively reveal underlying issues.

What is the best way to clarify project requirements before starting architecture decisions?

The best way to clarify requirements for architecture decisions is through a structured questioning session that gathers evidence and handles contradictions. This ensures accountability by focusing on a bedrock-focused summary.

How does evidence-driven questioning work during technical investigations?

Evidence-driven questioning works by applying a branching strategy and gathering evidence to manage multiple threads and verify assumptions. Contradiction detection ensures teams avoid surface-level explanations during technical investigations.

Can I use structured root-cause analysis for debugging complex systems?

Yes, you can use structured root-cause analysis for debugging complex systems. It guides deep investigative reasoning by keeping questions traceable to the initial objective, progressively revealing deeper causes one question at a time.

When should I use a one-question-at-a-time loop instead of general brainstorming?

You should use a one-question-at-a-time loop instead of general brainstorming when surface explanations are insufficient. It enforces objective seeding, contradiction handling, and evidence gathering to reach a bedrock-focused conclusion.