ClawHub
Back to skill

Security audit

hook-scope-guide

Security checks across malware telemetry and agentic risk

Overview

This is an informational guide for choosing Claude Code hook scope, with examples that are mostly purpose-aligned but should be used carefully when logging tool input.

Before copying the logging examples, consider whether tool input could include secrets, proprietary code, prompts, or sensitive file paths. Prefer redaction, minimal logging, restrictive file permissions, and clear retention limits, especially for global hooks that run across all sessions.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
86% confidence
Finding
The skill includes concrete logging examples that write tool input and session activity to disk, including values like $CLAUDE_TOOL_INPUT. Tool input may contain secrets, proprietary code, file paths, prompts, or other sensitive data, and the privacy risk is not clearly foregrounded where the examples are presented.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.