ClawHub
Back to skill

Security audit

Subagent Health Monitor

Security checks across malware telemetry and agentic risk

Overview

This skill is a local Claude Code monitoring tool that stores subagent health data on disk, with no evidence of network, credential, or destructive behavior.

Install only if you are comfortable with subagent monitoring metadata being kept locally under ~/.claude. Avoid putting secrets or sensitive customer details in task names or descriptions, and delete ~/.claude/subagent-health-state.json when you want to clear retained state.

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
87% confidence
Finding
The invocation examples use very broad phrases such as 'Check fleet health' and 'Is subagent-3 stuck?', which can overlap with normal conversation and may cause unintended skill activation. In an agent environment, ambiguous triggers can lead to unexpected tool execution, state changes, or disclosure of monitoring data when the user did not explicitly intend to invoke this skill.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill states that monitoring state is stored locally in ~/.claude/ but does not warn users about persistence, retention, or the sensitivity of operational metadata being recorded. Even if no API keys are used, persisted agent health data can reveal workflow details, task structure, timing, and potentially sensitive project activity across sessions.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.