ClawHub
Back to skill

Security audit

System Load Monitor

Security checks across malware telemetry and agentic risk

Overview

This skill is a local system-load checker whose sensitive behavior is mostly disclosed and proportionate, with a privacy caution around process details.

Install this only if you want the agent to run local system-load checks and possibly delay resource-heavy work. Do not share the JSON output outside your environment unless you are comfortable exposing process usernames and command lines, and require explicit approval before sending any alert through Feishu or another external service.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill clearly instructs the agent to execute local Python scripts and use shell access, but it declares no permissions. This creates a trust and governance gap: a reviewer or runtime may underestimate the skill's capabilities, and users may invoke it without realizing it can read system state and execute commands.

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding
The skill description claims automatic task pausing and resumption, but the documented behavior only checks load, emits recommendations, and exposes process information. This mismatch is dangerous because operators may rely on protections that do not actually exist, causing unsafe execution of resource-intensive tasks or unintended disclosure of process details.

Context-Inappropriate Capability

Low
Confidence
85% confidence
Finding
The script collects and returns usernames and full command lines for top processes, which can expose sensitive operational details such as service names, arguments, file paths, or embedded secrets passed on the command line. In the context of a monitoring skill, this exceeds the minimum data needed for load-threshold decisions and increases information disclosure risk if output is logged, surfaced to users, or sent to other agents.

Vague Triggers

Medium
Confidence
76% confidence
Finding
The activation criteria are broad phrases like needing to monitor server status or execute resource-intensive tasks, which can cause the skill to trigger in contexts where shell-based monitoring is unnecessary or inappropriate. In an agent environment, overbroad invocation increases the chance of unnecessary command execution and access to host telemetry.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.