What is the 'Iron Law' in this skill?

The Iron Law is a strict procedural requirement that no code fixes can be proposed or implemented until a root cause has been definitively identified and verified through evidence.

When does the skill escalate to the user?

The skill triggers an escalation when three or more fix attempts fail, as this typically indicates an underlying architectural problem rather than a simple logic bug.

How does it handle complex, multi-layered systems?

It traces data flow across component boundaries, logging inputs and outputs at each layer to isolate exactly where the failure occurs before investigating specific component code.

Does this skill require an existing test suite?

While it works best with existing suites, Phase 4 of the workflow explicitly requires the creation of a new failing test case to prove the bug exists and verify the fix.

Systematic Debugging Workflow

Name: Systematic Debugging Workflow
Author: thermiteau

bythermiteau

0•

보안 및 테스팅

Implements a disciplined four-phase approach to debugging that prioritizes root cause investigation over trial-and-error fixing.

The Systematic Debugging skill transforms how Claude handles technical failures by enforcing a rigorous 'Iron Law': no fixes without a verified root cause. It guides the AI through a structured methodology involving deep evidence gathering, pattern analysis against working examples, scientific hypothesis testing, and rigorous verification via failing tests. By preventing common anti-patterns like 'shotgun debugging' or masking issues with timeouts, this skill ensures that software problems are resolved at their source, reducing regressions and identifying deeper architectural weaknesses that require human escalation.

주요 기능

01Multi-component boundary analysis for isolating bugs in complex API and service stacks

02Structured 4-phase lifecycle: Investigation, Pattern Analysis, Hypothesis, and Verification

030 GitHub stars

04Automatic escalation protocol triggered after three failed fix attempts to signal architectural issues

05Requirement for reproduction test cases to prevent future regressions

06Mandatory root cause investigation phase before any code modifications

사용 사례

01Resolving persistent test failures that have resisted standard 'quick-fix' attempts

02Debugging unexpected behavior in unfamiliar codebases by performing comparative pattern analysis

03Troubleshooting intermittent or flaky bugs in distributed, multi-layered systems

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add thermiteau/maverick-private mav-systematic-debugging

For use in Claude.ai and ChatGPT

주요 기능

01Multi-component boundary analysis for isolating bugs in complex API and service stacks

02Structured 4-phase lifecycle: Investigation, Pattern Analysis, Hypothesis, and Verification

030 GitHub stars

04Automatic escalation protocol triggered after three failed fix attempts to signal architectural issues

05Requirement for reproduction test cases to prevent future regressions

06Mandatory root cause investigation phase before any code modifications

사용 사례

01Resolving persistent test failures that have resisted standard 'quick-fix' attempts

02Debugging unexpected behavior in unfamiliar codebases by performing comparative pattern analysis

03Troubleshooting intermittent or flaky bugs in distributed, multi-layered systems