ClawHub
Back to skill

Security audit

hook-auditor

Security checks across malware telemetry and agentic risk

Overview

The skill is a coherent hook-auditing tool, but it can inspect and help change global Claude configuration after broad activation triggers.

Install only if you want an agent to audit Claude Code hooks. Before running it, confirm that you want it to read ~/.claude configuration, and review any proposed cleanup carefully. It does require confirmation and backups before changes, but those changes can alter persistent Claude behavior.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
82% confidence
Finding
The trigger list is very broad and matches common complaints like slowness, token usage, or unexpected behavior, which can cause the skill to activate in many unrelated conversations. In this skill's context, activation leads to inspection of sensitive agent configuration under ~/.claude and potentially recommendations for global hook removal, so over-triggering increases the chance of unnecessary access and risky guidance.

Agent Config Directory Access

High
Category
Agent Snooping
Content
| 位置 | 是否生效 |
|---|---|
| `~/.claude/settings.json`、`settings.local.json` | **生效**（用户自己配的，默认可信） |
| `~/.claude/skills/<name>/hooks/hooks.json` | **生效**，安装即常驻 ← 意外 hook 主要来源 |
| `~/.claude/plugins/cache/<mkt>/<plugin>/hooks/hooks.json` | **仅当** settings 的 `enabledPlugins["<plugin>@<mkt>"]` 为 `true` |
| `~/.claude/plugins/marketplaces/**/hooks/hooks.json` | **永不生效**——marketplace 仓库缓存 ← 最易误判 |
Confidence
90% confidence
Finding
~/.claude/settings.json

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.