debugging

Reproduce software defects and verify fixes with evidence-based workflows.

1|1|Updated Dec 28, 2025

Search Tags:#debugging #troubleshooting #verification #evidence #reproduce #root-cause-analysis

Ready to run

npx skills add https://github.com/ssimhan/meal-planner --skill debugging-ssimhan

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: debugging
Source: https://github.com/ssimhan/meal-planner/tree/main/skills/debugging
Command: npx skills add https://github.com/ssimhan/meal-planner --skill debugging-ssimhan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps teams systematically identify, reproduce, and verify defects using evidence-based methods, reducing guesswork and patchwork fixes.

Core Features & Use Cases

Reproduce & trace: Establish exact reproduction steps and trace data flow to the root cause.
Evidence-driven fixes: Gather evidence, test hypotheses, and validate fixes with minimal risk.
Rigorous verification: Apply a red-green-verify cycle to ensure defects are resolved across components.

Quick Start

Start the debugging workflow to reproduce a bug, collect boundary data, and generate a minimal reproducible example.

Frequently Asked Questions about debugging

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

FAQPage Schema

How do I systematically find the root cause of a software bug?▼

To find a root cause systematically, you reproduce the bug to build an evidence base, hypothesize the defect source, test with minimal failing cases, and verify the fix. This disciplined workflow eliminates guesswork by relying on collected trace data.

What is the best way to reproduce and trace unexpected software behavior?▼

The best way to reproduce unexpected behavior is establishing exact reproduction steps and tracing data flow across components. This disciplined approach builds a solid evidence base for testing hypotheses and applying rigorous verification to resolve defects.

How do I verify a bug fix without introducing regressions across components?▼

To verify a bug fix without regressions, apply a red-green-verify cycle using minimal failing tests. This ensures the software defect is resolved across components by documenting confirmatory results and validating evidence before finalizing.

When do I need an evidence-led debugging workflow for troubleshooting test failures?▼

You need an evidence-led debugging workflow when troubleshooting test failures requires systematic root-cause analysis rather than patchwork fixes. It enforces disciplined defect reproduction, hypothesis testing, and rigorous verification to build a solid evidence base.

Can I use this debugging workflow for projects with minimal failing tests?▼

Yes, you can use this debugging workflow for projects with minimal failing tests. It generates minimal reproducible examples and collects boundary data to establish an evidence base, making it suitable for projects facing unexpected behavior.