When should I use the problem-analysis skill?

Use it for complex bugs, intermittent errors, or unexpected system behavior where the cause isn't immediately obvious and requires evidence-based investigation.

How does the confidence assessment work?

It uses a point-based scoring system evaluating specific evidence, alternative hypotheses, symptom explanation, and logical framing to determine if a diagnosis is ready for a fix.

What is the Problem Analysis skill?

It is a Claude Code extension that guides the AI through a structured methodology to find the underlying cause of a bug before proposing a fix, ensuring the actual problem is understood.

Does this work with any programming language?

Yes, the methodology is language-agnostic and focuses on logical investigation patterns and evidence gathering applicable to any software stack.

Why doesn't this skill suggest solutions?

By focusing exclusively on the root cause, it prevents 'solution bias' where a developer might accidentally fix a symptom rather than the actual condition causing the failure.

Problem Analysis & Root Cause Diagnostic

Name: Problem Analysis & Root Cause Diagnostic
Author: alphaleonis

byalphaleonis

0•

개발자 도구

Performs structured, evidence-based investigations to identify technical root causes without jumping to premature solutions.

The Problem Analysis skill is a specialized diagnostic framework for Claude Code designed to eliminate guesswork and confirmation bias during the debugging process. By enforcing a rigorous five-phase workflow—Gate, Hypothesize, Investigate, Formulate, and Output—it helps developers isolate the exact conditions causing a system failure. Instead of suggesting quick fixes that might only mask symptoms, this skill focuses on defining observable conditions and validating them with specific code or log evidence, ensuring a deep understanding of issues before any implementation begins.

주요 기능

01Standardized diagnostic reporting for clear hand-off to implementation phases

02Evidence-based confidence scoring to quantify diagnostic accuracy

03Multi-hypothesis testing to prevent confirmation bias and tunnel vision

04Root cause framing that focuses on observable conditions rather than missing features

050 GitHub stars

06Structured five-phase investigation workflow to ensure thoroughness

사용 사례

01Diagnosing unexpected behavior in complex distributed systems or middleware

02Investigating intermittent system failures or edge cases without obvious causes

03Performing a pre-fix audit to ensure the proposed solution addresses the true root cause

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add alphaleonis/decaf-claude-config problem-analysis

For use in Claude.ai and ChatGPT

주요 기능

01Standardized diagnostic reporting for clear hand-off to implementation phases

02Evidence-based confidence scoring to quantify diagnostic accuracy

03Multi-hypothesis testing to prevent confirmation bias and tunnel vision

04Root cause framing that focuses on observable conditions rather than missing features

050 GitHub stars

06Structured five-phase investigation workflow to ensure thoroughness

사용 사례

01Diagnosing unexpected behavior in complex distributed systems or middleware

02Investigating intermittent system failures or edge cases without obvious causes

03Performing a pre-fix audit to ensure the proposed solution addresses the true root cause

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add alphaleonis/decaf-claude-config problem-analysis

For use in Claude.ai and ChatGPT