debug-claude

Diagnose Claude Code hook failures and session anomalies from debug logs.

Updated May 7, 2026
One-click install
npx skills add https://github.com/tmalcolm-0607/mad-council-claw --skill debug-claude
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: debug-claude
Source: https://github.com/tmalcolm-0607/mad-council-claw/tree/main/.claude/skills/debug-claude
Command: npx skills add https://github.com/tmalcolm-0607/mad-council-claw --skill debug-claude

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates the guesswork when Claude Code behaves unexpectedly, providing structured diagnostics for hook failures, session anomalies, agent team issues, and file recovery needs.

Core Features & Use Cases

  • Hook Diagnostics: Identify why PreToolUse or PostToolUse hooks fail silently or produce incorrect permission decisions.
  • Session Analysis: Review tool usage patterns, errors, and warnings across Claude Code debug logs.
  • Agent Team Analysis: Evaluate teammate performance, token usage, and warnings in multi-agent sessions.
  • File Recovery: Restore files from session transcripts after destructive operations.
  • Health Checks: Run quick aggregate assessments to determine if the environment is stable.

Quick Start

Use the debug-claude skill to diagnose why your PreToolUse hook is failing silently and recover any lost files from the most recent session.

Frequently Asked Questions about debug-claude

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

FAQPage Schema
How do I diagnose Claude Code hook failures that are silently producing incorrect permission decisions?

Diagnosing Claude Code hook failures involves analyzing debug logs and settings files within $CLAUDE_CONFIG_DIR to identify why PreToolUse or PostToolUse hooks fail silently and produce incorrect permission decisions.

Can I recover files lost after destructive operations using Claude Code session transcripts?

File recovery from Claude Code session transcripts is possible by parsing project JSONL transcripts to restore files lost after destructive operations, eliminating the guesswork when unexpected behavior occurs.

What is the best way to analyze agent team performance and token usage in multi-agent Claude Code sessions?

Analyzing agent team performance in Claude Code involves evaluating teammate token usage, performance metrics, and warnings across debug logs and session transcripts in multi-agent environments.

Does Claude Code debugging require access to specific environment directories and log formats?

Claude Code debugging requires access to $CLAUDE_CONFIG_DIR debug logs, project JSONL transcripts, and settings files to perform grep-based analysis and health checks across CLI, Electron, or headless environments.

Why is my Claude Code session experiencing anomalies and how do I run a health check?

Session anomalies in Claude Code are diagnosed by running health checks that perform quick aggregate assessments on debug logs and tool usage patterns to determine if the environment is stable.

Can I debug Claude Code issues in headless environments or only in the CLI and Electron app?

Debugging Claude Code issues applies to CLI, Electron app, and headless environments, using grep-based analysis on debug logs and session transcripts to diagnose operational problems consistently.