ClawHub
Back to skill

Security audit

Watchdog - Open claw - Mac

Security checks across malware telemetry and agentic risk

Overview

This is a coherent local watchdog, but it needs Review because it can automatically kill and restart local services, edit user config, and send alerts externally with limited disclosure.

Use this only on a machine where it is acceptable for the watchdog to restart the intended llama.cpp/OpenClaw services. Before unattended use, verify the process match targets, consider DRY_RUN, disable Telegram or provide tokens only intentionally, and back up any Pi/OpenClaw config it may rewrite.

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 (3)

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill description states it is a 'self-healing supervisor' that auto-restarts on failure, but it does not clearly warn users that the watchdog may terminate and relaunch local inference server processes. In an agent/plugin context, process-control behavior can affect availability, interrupt active workloads, or restart the wrong local service if misconfigured, so omitting an explicit warning meaningfully reduces informed consent and safe deployment.

External Transmission

Medium
Category
Data Exfiltration
Content
local bot_token="${TELEGRAM_BOT_TOKEN:-}"
  local chat_id="${TELEGRAM_CHAT_ID:-}"
  [[ -z "$bot_token" || -z "$chat_id" ]] && return 0
  curl -s -X POST "https://api.telegram.org/bot${bot_token}/sendMessage" \
    -d "chat_id=${chat_id}" \
    -d "text=🔔 Watchdog: ${title} — ${msg}" \
    -d "disable_notification=true" >/dev/null 2>&1 || true
Confidence
65% confidence
Finding
curl -s -X POST "https://api.telegram.org/bot${bot_token}/sendMessage" \ -d "chat_id=${chat_id}" \ -d "text=🔔 Watchdog: ${title} — ${msg}" \ -d

External Transmission

Medium
Category
Data Exfiltration
Content
local bot_token="${TELEGRAM_BOT_TOKEN:-}"
  local chat_id="${TELEGRAM_CHAT_ID:-}"
  [[ -z "$bot_token" || -z "$chat_id" ]] && return 0
  curl -s -X POST "https://api.telegram.org/bot${bot_token}/sendMessage" \
    -d "chat_id=${chat_id}" \
    -d "text=🔔 Watchdog: ${title} — ${msg}" \
    -d "disable_notification=true" >/dev/null 2>&1 || true
Confidence
65% confidence
Finding
https://api.telegram.org/

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.