ClawHub
Back to skill

Security audit

WSL-PowerShell Controller

Security checks across malware telemetry and agentic risk

Overview

This skill appears to do what it says, but it gives an agent broad Windows PowerShell control over the Windows host from WSL without strong guardrails.

Install only if you intentionally want your agent to control Windows PowerShell from WSL. Prefer the reviewed ClawHub package over live GitHub downloads, review commands before they run, avoid untrusted scripts, and require explicit approval for file changes, process control, program launches, execution-policy bypasses, or administrator prompts.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The README states the skill will auto-load and that agents can automatically use it based on user requests, but it does not define meaningful guardrails, approval requirements, or scope limits. In a skill that bridges WSL to Windows PowerShell, overly broad activation increases the chance an agent invokes powerful host-control functionality unintentionally or in response to ambiguous prompts.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The usage examples include system-affecting actions such as copying files, creating files, starting programs, and forcefully stopping processes, but the nearby examples do not prominently warn about side effects or the need for confirmation. Because this skill is designed for agent consumption and controls Windows from WSL, these examples normalize impactful operations that an agent or user may reuse without appreciating the risk.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill includes a forceful process-termination example (`Stop-Process -Name notepad -Force`) without clearly warning that the same pattern can abruptly kill arbitrary applications, causing data loss or service disruption. In a WSL-to-Windows control skill, this is more dangerous than a normal documentation snippet because it explicitly enables cross-environment control of host-side Windows processes from Linux.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.