ClawHub
Back to skill

Security audit

Rdk X5 Monitor

Security checks across malware telemetry and agentic risk

Overview

This is a mostly read-only RDK X5 hardware monitoring skill, with a few visible troubleshooting suggestions that can change system state if a user chooses to run them.

Install if you want an agent to inspect RDK X5 hardware status. Do not let the agent run kill or sudo apt clean unless you explicitly want to stop a process or clear package cache and have reviewed the target command.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
This skill is explicitly positioned as a read-only hardware monitoring skill, but the troubleshooting section instructs users to kill processes and perform cleanup actions. That expands the skill from passive observation into operational system modification, creating a mismatch that can cause an agent or user to take destructive actions under the guise of monitoring.

Description-Behavior Mismatch

Medium
Confidence
97% confidence
Finding
The documented behavior exceeds the declared scope of a monitoring-only skill by recommending actions that alter system state, such as terminating processes and running `sudo apt clean`. In an agent setting, this kind of scope creep is dangerous because users may invoke a supposedly safe read-only skill and still be led into impactful administrative changes.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal