debug

Diagnoses software bugs using evidence-first root cause analysis and verified fixes.

7|2|Updated Dec 10, 2025
One-click install
npx skills add https://github.com/sbusso/claude-system --skill debug-sbusso
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: debug
Source: https://github.com/sbusso/claude-system/tree/main/skills/debug
Command: npx skills add https://github.com/sbusso/claude-system --skill debug-sbusso

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The Debug skill provides a structured, evidence-first methodology to find, explain, and fix software bugs, reducing time spent chasing symptoms and preventing regressions.

Core Features & Use Cases

  • Evidence Gathering: Collect full error messages, stack traces, affected components, and recent diffs to build a reproducible case.
  • Root Cause Analysis: Form and test focused hypotheses one variable at a time, tracing dependencies to the source.
  • Verified Solutions: Implement minimal fixes, validate against reproduction steps, and check for regressions before claiming resolution.
  • Use Case: Troubleshoot a failing unit test after a recent refactor by reproducing the failure, isolating the change, and applying a minimal patch with verification.

Quick Start

Ask the debug skill to gather the error details, propose 2-3 testable hypotheses, and recommend a minimal verified fix for the failing test.

Frequently Asked Questions about debug

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

FAQPage Schema
How do I find the root cause of a runtime exception using a stack trace?

To find the root cause of a runtime exception, gather the full stack trace and recent code diffs, then test focused hypotheses one variable at a time to trace dependencies back to the source. This isolates the exact failure point.

What's the best way to troubleshoot a failing unit test after a code refactor?

The best way to troubleshoot a failing unit test is to reproduce the failure consistently, isolate the recent changes, and apply a minimal verified patch. Validate the fix against reproduction steps to prevent regressions.

How does hypothesis-driven debugging work for integration issues?

Hypothesis-driven debugging works by forming 2-3 testable hypotheses about the integration issue, checking each one against collected evidence like error outputs and stack traces, and discarding disproven variables until the root cause is found.

Do I need reproducible error outputs to fix software bugs effectively?

Yes, you need reproducible error outputs to fix software bugs effectively. An evidence-first debugging process requires consistent reproduction steps to validate minimal fixes and check for regressions before claiming resolution.

Why does chasing symptoms fail to resolve runtime exceptions during code review?

Chasing symptoms fails to resolve runtime exceptions because it skips disciplined root cause analysis. Without testing focused hypotheses against stack traces and recent diffs, the underlying fault remains unaddressed and causes regressions.