Skill

hooks-health

Run a comprehensive health check of the Claude Code hook system.

Install

Run in your terminal

npx claudepluginhub revpalsfdc/opspal-commercial --plugin opspal-core

Tool Access

This skill is limited to using the following tools:

ReadBashGlobGrep

Skill Content

Similar Skills

agent-payment-x402

Enables AI agents to execute x402 payments with per-task budgets, spending controls, and non-custodial wallets via MCP tools. Use when agents pay for APIs, services, or other agents.

ecc

140.7k

agent-eval

Compares coding agents like Claude Code and Aider on custom YAML-defined codebase tasks using git worktrees, measuring pass rate, cost, time, and consistency.

ecc

140.7k

agent-harness-construction

Designs and optimizes AI agent action spaces, tool definitions, observation formats, error recovery, and context for higher task completion rates.

ecc

140.7k

Stats

Parent Repo Stars0

Parent Repo Forks1

Last CommitMar 18, 2026

Actions

View Source View Plugin View on GitHub View README

Usage

/hooks-health # Full diagnostic (terminal output) /hooks-health --quick # Skip execution tests (faster) /hooks-health --verbose # Include detailed fix instructions /hooks-health --format json # JSON output for automation /hooks-health --format markdown # Markdown report

What It Checks (10-Stage Pipeline)

Stage

Check

Detects

Configuration Discovery

Missing hooks.json, invalid JSON

File Permissions

Scripts exist, executable

Syntax Validation

bash -n, node --check errors

Dependency Detection

Missing jq, node, bc, etc.

Execution Test

Timeout, crash, non-zero exit

Silent Failure Detection

Empty output, missing fields, hidden errors

Circuit Breaker State

Open circuits, high failure counts

Log Analysis

Recent errors, patterns

Cross-Reference

Orphaned scripts, duplicates

Health Summary

Score, recommendations

Code

Status

Meaning

HEALTHY

All checks pass, score 90-100

DEGRADED

Warnings present, score 50-89

UNHEALTHY

Critical issues, score 0-49

Execution

Run the health checker, then auto-submit a reflection if issues are found:

PLUGIN_ROOT="${CLAUDE_PLUGIN_ROOT:-$(dirname $(dirname $(dirname $0)))}" CHECKER="$PLUGIN_ROOT/scripts/lib/hook-health-checker.js" GENERATOR="$PLUGIN_ROOT/scripts/lib/hook-health-reflection-generator.js" # Run health check normally (user sees terminal output) node "$CHECKER" "$@" EXIT_CODE=$? # If degraded/unhealthy, generate and submit reflection in background if [ $EXIT_CODE -ne 0 ] && [ -f "$GENERATOR" ]; then node "$CHECKER" --format json --quick 2>/dev/null | node "$GENERATOR" --stdin >/dev/null 2>&1 & disown fi exit $EXIT_CODE

Hook System Health Check

Run comprehensive diagnostics on all configured hooks with silent failure detection.

Quick Start

/hooks-health

Usage

/hooks-health                    # Full diagnostic (terminal output)
/hooks-health --quick            # Skip execution tests (faster)
/hooks-health --verbose          # Include detailed fix instructions
/hooks-health --format json      # JSON output for automation
/hooks-health --format markdown  # Markdown report

What It Checks (10-Stage Pipeline)

Stage	Check	Detects
1	Configuration Discovery	Missing hooks.json, invalid JSON
2	File Permissions	Scripts exist, executable
3	Syntax Validation	bash -n, node --check errors
4	Dependency Detection	Missing jq, node, bc, etc.
5	Execution Test	Timeout, crash, non-zero exit
6	Silent Failure Detection	Empty output, missing fields, hidden errors
7	Circuit Breaker State	Open circuits, high failure counts
8	Log Analysis	Recent errors, patterns
9	Cross-Reference	Orphaned scripts, duplicates
10	Health Summary	Score, recommendations

Silent Failure Detection

Detects hooks that:

Exit 0 but produce no output when output expected
Return JSON without required fields (systemMessage, decision)
Contain error messages but don't fail
Near-timeout without reporting

Exit Codes

Code	Status	Meaning
0	HEALTHY	All checks pass, score 90-100
1	DEGRADED	Warnings present, score 50-89
2	UNHEALTHY	Critical issues, score 0-49

Execution

Run the health checker, then auto-submit a reflection if issues are found:

PLUGIN_ROOT="${CLAUDE_PLUGIN_ROOT:-$(dirname $(dirname $(dirname $0)))}"
CHECKER="$PLUGIN_ROOT/scripts/lib/hook-health-checker.js"
GENERATOR="$PLUGIN_ROOT/scripts/lib/hook-health-reflection-generator.js"

# Run health check normally (user sees terminal output)
node "$CHECKER" "$@"
EXIT_CODE=$?

# If degraded/unhealthy, generate and submit reflection in background
if [ $EXIT_CODE -ne 0 ] && [ -f "$GENERATOR" ]; then
  node "$CHECKER" --format json --quick 2>/dev/null | node "$GENERATOR" --stdin >/dev/null 2>&1 &
  disown
fi

exit $EXIT_CODE