debug

Diagnoses OMC or Claude Code session issues using logs, traces, and state inspection.

Updated Apr 17, 2026
One-click install
npx skills add https://github.com/dropsyoon/oh-my-claudecode --skill debug-dropsyoon
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: debug
Source: https://github.com/dropsyoon/oh-my-claudecode/tree/main/skills/debug
Command: npx skills add https://github.com/dropsyoon/oh-my-claudecode --skill debug-dropsyoon

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The debug skill helps you diagnose confusing OMC/Claude-Code session problems by turning unclear symptoms into actionable evidence and next steps.

Core Features & Use Cases

  • Evidence-first diagnosis: prioritize real traces, state, and relevant local artifacts over guesses.
  • Targeted reproduction: narrow the failing behavior to isolate the true failure signal.
  • Orchestration-aware guidance: focus on trace/state surfaces when issues involve hooks, agent flow, or orchestration breakage.

Quick Start

Ask the skill to diagnose your current OMC session failure by reviewing the latest logs and tracing the steps until the first observable failure, then propose the smallest next verification action.

Frequently Asked Questions about debug

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

FAQPage Schema
How do I debug Claude Code orchestration failures and hook issues?

Debug Claude Code orchestration failures by applying an evidence-first workflow that inspects logs, traces, and state to isolate true failure signals in hooks and agent-flow breakage, separating symptoms from root cause.

What is the best way to diagnose confusing OMC session runtime behavior?

The best way to diagnose confusing OMC session runtime behavior is targeted reproduction that narrows the failing behavior. This approach outputs an observed failure, a root-cause hypothesis with supporting evidence, and the smallest next action.

Can I trace agent flow and state to find the root cause of workflow breakage?

Yes, you can trace agent flow and state to find the root cause of workflow breakage. The diagnosis focuses on trace and state surfaces when issues involve orchestration, ensuring root-cause hypotheses are backed by supporting evidence.

Does this debugging approach work for active development sessions or post-mortem analysis?

This debugging approach works specifically for active development sessions where confusing runtime behavior occurs. It requires an evidence-first workflow and trace/state inspection to output the smallest next verification action.

Why does my Claude Code session fail without clear error messages?

Your Claude Code session may fail without clear error messages due to orchestration or hook-flow breakage. Diagnosing the issue requires prioritizing real traces, state, and relevant local artifacts over guesses to pinpoint the true failure signal.