debugging-methodology

Diagnose unexplained software failures through an evidence-first root-cause workflow.

7|2|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/AndurilCode/craftwork --skill debugging-methodology-andurilcode
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: debugging-methodology
Source: https://github.com/AndurilCode/craftwork/tree/main/skills/debugging-methodology
Command: npx skills add https://github.com/AndurilCode/craftwork --skill debugging-methodology-andurilcode

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you debug software failures you cannot explain—such as failing tests, unexpected behavior, regressions, or intermittent bugs—so you can isolate the true root cause instead of guess-and-patching.

Core Features & Use Cases

  • Evidence-first failure analysis: Reproduce the issue and collect exact errors, logs, system state, and recent changes before proposing explanations.
  • Hypothesis ranking and fast elimination: Generate multiple plausible causes, rank them by likelihood, testability, and impact, and run quick checks to narrow the search.
  • Binary isolation and verification: Isolate the breaking dimension using bisecting/stubbing/minimization strategies, then confirm mechanism-level root cause and validate with regression tests.

Use case: A CI test started failing after a recent refactor, and you suspect a code change but cannot tell which assumption broke; this skill structures the investigation from reproduction through a root-cause fix and regression coverage.

Quick Start

Apply the debugging-methodology skill to diagnose your broken test or intermittent failure by providing the exact failure symptom, logs, and what changed since it last worked.

Frequently Asked Questions about debugging-methodology

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

FAQPage Schema
How do I find the root cause of intermittent test failures instead of just patching symptoms?

To find the root cause of intermittent test failures, apply an evidence-first workflow that reproduces the issue, collects logs, ranks hypotheses by testability, and isolates the breaking dimension via binary search before writing a regression test.

What is the best way to debug a regression introduced by a recent code refactor?

Debugging a regression requires capturing exact error messages and recent changes, then using binary search over commits or code to isolate the broken assumption, confirming the causal mechanism with a targeted regression test.

How does git bisect work for isolating environment-dependent bugs?

Git bisect isolates environment-dependent bugs by performing a binary search over commits, systematically narrowing down the exact code change that introduced the failure until the root cause is identified and verified.

Why do my CI tests fail unexpectedly after a refactor when I cannot tell which assumption broke?

Unexpected CI test failures occur when hidden assumptions break during refactoring; resolve them by generating multiple plausible hypotheses, ranking them by likelihood, and running quick checks to isolate the true causal mechanism.

Can I use hypothesis testing to diagnose unexpected runtime behavior without guessing the fix?

Yes, hypothesis testing diagnoses unexpected runtime behavior by ranking multiple plausible causes by likelihood and impact, then running quick checks to systematically eliminate incorrect assumptions before attempting a fix.

When should I not use a structured debugging methodology for software failures?

A structured debugging methodology is unnecessary when the failure is already understood or trivially reproducible, but essential for unexplained regressions, intermittent bugs, and environment-dependent failures requiring mechanism-level verification.