diagnose

Diagnose Claude Code skill and hook failures via a four-phase root-cause protocol.

1|Updated Apr 1, 2026
One-click install
npx skills add https://github.com/charleshall888/cortex-command --skill diagnose-charleshall888
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: diagnose
Source: https://github.com/charleshall888/cortex-command/tree/main/plugins/cortex-core/skills/diagnose
Command: npx skills add https://github.com/charleshall888/cortex-command --skill diagnose-charleshall888

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Systematic debugging and root-cause analysis for Claude Code skills, hooks, lifecycle, and overnight runner issues, reducing guesswork and speeding fixes.

Core Features & Use Cases

  • Four-phase debugging protocol (Phase 1 through Phase 4) to identify and verify root causes.
  • Lifecycle and escalation guidance, including team-based investigations when needed.
  • Debug-session artifact guidance and reference material for structured post-mortems.

Quick Start

Invoke /cortex-core:diagnose on a failing task and follow the four-phase debugging protocol.

Frequently Asked Questions about diagnose

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

FAQPage Schema
How do I debug a failing Claude Code skill systematically?▼

To debug a failing Claude Code skill systematically, you can apply a four-phase debugging protocol that identifies and verifies root causes through structured artifact tracking and lifecycle analysis, reducing guesswork and speeding fixes.

What is the four-phase debugging protocol for Claude Code lifecycle tasks?▼

The four-phase debugging protocol for Claude Code lifecycle tasks is a structured root-cause analysis process spanning Phase 1 through Phase 4, designed to guide fixes with disciplined process steps and debug-session artifacts.

Can I run Claude Code debugging sessions autonomously without manual input?▼

Yes, you can run Claude Code debugging sessions autonomously or interactively, applying the structured debugging protocol across phases with explicit escalation paths for team-based investigations when needed.

How do I track debug artifacts for a Claude Code hook issue?▼

To track debug artifacts for a Claude Code hook issue, follow the debug-session artifact guidance provided by the systematic debugging process, which uses references and artifacts to guide fixes and structure post-mortems.

When should I escalate a Claude Code debugging session to a team investigation?▼

You should escalate a Claude Code debugging session to a team investigation when the structured four-phase protocol identifies complex lifecycle issues that require explicit escalation paths for collaborative root-cause resolution.

Does systematic debugging work for overnight runner issues in Claude Code?▼

Yes, systematic debugging works for overnight runner issues in Claude Code by applying root-cause analysis and lifecycle guidance to resolve failures through structured artifact tracking and disciplined process steps.