System Watchdog

• Outcome: Review
• Latest audit: May 28, 2026 at 8:40 AM UTC
• Version:
  2.0.0

Overview

This is a coherent local system-monitoring skill, but it stores full process command lines persistently, which can expose secrets or private paths.

Review before installing if your machine may run commands with secrets in arguments. The skill appears legitimate and local-only, but its state file can preserve sensitive command-line data; keep the state file in a protected location and avoid passing passwords, API keys, or tokens on process command lines.

SkillSpector

By NVIDIA

Vulnerability Patterns

• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 86% confidence
• Finding: The skill instructs the agent to execute a shell script, read and write a persistent state file, and use environment-variable-controlled paths, but no permissions are declared to make those capabilities explicit. This creates a mismatch between the documented behavior and the security model, increasing the chance of unintended shell execution or filesystem access without proper review or containment.

VirusTotal

66/66 vendors flagged this skill as clean.