ClawHub
Back to skill

Security audit

Health Guardian

Security checks across malware telemetry and agentic risk

Overview

Health Guardian appears purpose-built for local health monitoring, but it needs review because it handles sensitive health data while under-disclosing cloud sync, Telegram-style alerting, and ongoing scheduled processing risks.

Review before installing. Use this only if you are comfortable letting it read Apple Health exports and retain raw health telemetry locally. Verify the exact source path, protect or encrypt the data directory, avoid external alert channels unless the user and recipients explicitly consent, and enable hourly cron only if you know how to disable it.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
86% confidence
Finding
The skill documentation describes reading from local iCloud/Apple Health export paths and writing health data into local JSON files, but it does not declare corresponding permissions. Undeclared file access is a security and trust problem because agents or users cannot accurately assess what local data the skill can read and persist, especially when the data is highly sensitive health information.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill promotes proactive health alerts and caregiver/agent notifications without clearly warning that alerts may contain protected or highly sensitive health information. In a health-monitoring context, omission of privacy and disclosure warnings increases the risk of accidental exposure, over-sharing, or transmission through insecure channels.

Missing User Warnings

High
Confidence
97% confidence
Finding
The example configuration sets Telegram as an alert channel for anomaly notifications without warning that this may transmit sensitive health data off-device to a third-party service. Because the skill handles chronic-condition and disability-related data, sending alerts through Telegram could expose highly sensitive information to unauthorized recipients, cloud retention, or account compromise.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.