Why can't I just apply a quick fix with this skill?

Quick patches often mask underlying issues and lead to technical debt. This skill enforces the 'Iron Law' of debugging: no fixes are permitted until the root cause is proven through evidence.

When should I use the Systematic Debugging skill?

Use it for any technical issue including test failures, production bugs, build errors, performance problems, or any unexpected behavior where the cause isn't immediately certain.

What happens if three or more fixes fail?

The skill triggers an Architectural Questioning phase. Instead of attempting a fourth fix, it prompts a discussion about whether the underlying pattern or system design is fundamentally flawed.

Does this skill help with complex distributed systems?

Yes, it includes specific protocols for 'Phase 1: Root Cause Investigation' that involve adding diagnostic instrumentation across component boundaries like APIs, build scripts, and databases.

How does this integrate with unit testing?

In the final Implementation phase, the skill requires creating a minimal failing test case before applying the fix, ensuring the solution is verified and preventing future regressions.

Systematic Debugging

Name: Systematic Debugging
Author: mbcoalson

bymbcoalson

0•

セキュリティとテスト

Enforces a rigorous four-phase debugging protocol to identify root causes and implement permanent fixes rather than surface-level patches.

Systematic Debugging is a specialized Claude Code skill designed to eliminate 'guess-and-check' programming by mandating a root-cause-first approach. It guides the AI through four distinct phases—Investigation, Pattern Analysis, Hypothesis Testing, and Implementation—ensuring that every bug, performance bottleneck, or test failure is thoroughly understood before any code is modified. This skill is particularly valuable during high-pressure emergencies or when multiple previous fix attempts have failed, helping developers break out of 'thrashing' cycles by identifying underlying architectural flaws rather than just treating symptoms.

主な機能

01Automatic escalation to architectural review after three failed fix attempts

020 GitHub stars

03Mandatory root cause investigation phase before any fix proposal

04Multi-component diagnostic instrumentation for complex system boundaries

05Backward data-flow tracing to find original error triggers

06Integrated Test-Driven Development (TDD) workflow for fix verification

ユースケース

01Resolving intermittent race conditions or environment-specific failures in CI/CD pipelines

02Debugging deep call-stack errors in multi-layered API or microservice architectures

03Breaking out of 'fix thrashing' loops where one change causes regressions elsewhere

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add mbcoalson/skills systematic-debugging

For use in Claude.ai and ChatGPT

Download Skill