What happens if a bug persists after multiple fix attempts?

If three or more fixes fail, the skill stops the cycle and triggers an architectural review to determine if the underlying pattern is fundamentally flawed.

How does this skill handle multi-component system failures?

It uses diagnostic instrumentation to log data at every component boundary, allowing you to see exactly where data flow breaks before investigating a specific component.

What is the 'Iron Law' of this debugging skill?

The Iron Law dictates that no fixes may be proposed until a Phase 1 Root Cause Investigation is completed, ensuring the underlying issue is understood before code is changed.

Is this approach slower than just trying a quick fix?

While it requires more initial steps, it is significantly faster than 'thrashing' (trying multiple unverified fixes), resulting in a 95% first-time fix rate.

Does this skill help with writing tests?

Yes, it requires the creation of a failing test case in the final phase to verify the fix and ensure the bug does not return.

Systematic Debugging Framework

Name: Systematic Debugging Framework
Author: shekohex

byshekohex

•

安全与测试

Enforces a rigorous four-phase debugging workflow to identify root causes and ensure reliable software fixes before attempting implementation.

The Systematic Debugging skill transforms Claude into a disciplined troubleshooter by mandating an 'Iron Law': no fixes without root cause investigation. By guiding the AI through structured phases of investigation, pattern analysis, hypothesis testing, and implementation, it eliminates time-wasting 'guess-and-check' behavior. This skill is essential for resolving complex integration issues, intermittent test failures, and architectural flaws that surface as recurring bugs, ensuring that every fix addresses the source rather than just the symptoms.

主要功能

01Diagnostic instrumentation for tracing data flow across multi-component systems

022 GitHub stars

03Mandatory root cause verification before any code modification is proposed

04Integration with Test-Driven Development (TDD) to create reproduction cases

05Strict four-phase debugging framework: Investigation, Pattern Analysis, Hypothesis, and Implementation

06Architectural sanity checks triggered automatically after three failed fix attempts

使用场景

01Debugging complex data flow issues across multiple API boundaries or service layers

02Troubleshooting legacy codebases where quick patches have historically introduced new regressions

03Resolving intermittent test failures in CI/CD pipelines that lack clear error origins

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add shekohex/dotai systematic-debugging

For use in Claude.ai and ChatGPT

Download Skill