debug

Investigate software bugs through a three-phase workflow and generate diagnostic debug reports.

23|2|Updated Mar 7, 2026
One-click install
npx skills add https://github.com/kirich1409/krozov-ai-tools --skill debug-kirich1409
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: debug
Source: https://github.com/kirich1409/krozov-ai-tools/tree/main/plugins/developer-workflow/skills/debug
Command: npx skills add https://github.com/kirich1409/krozov-ai-tools --skill debug-kirich1409

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Diagnoses issues by performing a structured root-cause investigation and generating a diagnostic debug report without applying any fixes.

Core Features & Use Cases

  • Structured, multi-phase root-cause investigation workflow (Phase 1: Root Cause Investigation, Phase 2: Binary Search Narrowing, Phase 3: Hypothesis and Confirmation)
  • Delegation to the debugging-expert agent to perform read-only analysis and evidence gathering
  • Produces a formal Debug Report with symptom, reproduction steps, evidence trail, and recommended fix direction
  • Escalation and artifact creation (swarm-report) for hand-off to the implement stage

Quick Start

Describe the bug symptoms and reproduction steps, then invoke the debug skill to generate a root-cause report.

Frequently Asked Questions about debug

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

FAQPage Schema
How do I perform a structured root-cause investigation for a software bug?

A structured root-cause investigation systematically traces symptoms to evidence using a multi-phase workflow. It analyzes software bugs, crashes, and test failures without applying fixes, producing a formal diagnostic report with reproduction steps and a recommended fix direction.

What is the best way to diagnose test failures and crashes without modifying the codebase?

Diagnosing test failures without modifying code is achieved through read-only analysis and evidence gathering. A debugging expert agent performs binary search narrowing and hypothesis confirmation to isolate the root cause, ultimately generating a diagnostic report instead of applying patches.

How do I narrow down the root cause of unexpected behavior in production environments?

Narrowing down unexpected behavior in production requires binary search narrowing to isolate variables. The investigation workflow progresses from initial symptom analysis through hypothesis confirmation, ensuring every conclusion is backed by a concrete evidence trail before documenting the root cause.

Does the debugging workflow stop after identifying the root cause of a bug?

Yes, the debugging workflow stops at diagnosis and does not apply any fixes. It generates a final root-cause artifact and a swarm-report for escalation, explicitly recommending a fix direction for hand-off to the implementation stage rather than modifying code directly.

Can I use this diagnosis process for both development and production environments?

Yes, the diagnosis process applies to unexpected behavior across both development and production environments. It handles software bugs, crashes, and test failures uniformly by guiding the investigation from symptom identification to evidence collection within its three-phase workflow.