Security audit

Desktop Control (Windows)

Security checks for vulnerabilities and agentic risk

Overview

This skill openly provides Windows desktop control, but its reach is broad enough that users should review it before installing.

Install only if you intentionally want an agent to control your Windows desktop. Keep it limited to explicit tasks, confirm destructive actions, avoid using screenshot or clipboard commands around sensitive data, and do not let it install VS Code extensions unless you have independently verified the publisher and extension.

A.I.G

Vulnerability Patterns

Insecure DependenciesIntroduces malicious components through unsafe dependency sources
Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally

Findings (1)

T08 · Insecure Dependencies

Warning

Location: scripts/vscode-control.ps1.txt:164
Finding: Unvalidated Force-Installation of Arbitrary VS Code Extensions
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

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

Findings (3)

Vague Triggers

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The skill description is extremely broad and invites activation for almost any request involving applications, windows, typing, files, processes, clipboard, or screen access. In an agentic system, this can cause over-selection of a high-privilege skill, increasing the chance of unintended desktop control, data exposure, or destructive actions from ambiguous user prompts.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The skill exposes screenshot capture and clipboard reading as routine actions without prominently warning that they may collect passwords, tokens, personal messages, or other sensitive on-screen data. Because this skill controls the live desktop, these capabilities are more dangerous here than in a passive informational context and can easily lead to privacy breaches if invoked casually.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: This script directly invokes keyboard and mouse simulation primitives against the currently focused desktop session, including arbitrary text entry, shortcut dispatch, clicks, and double-clicks, without any confirmation, target validation, or policy restrictions. In the context of a desktop-control skill, that enables destructive or security-sensitive actions such as closing applications, approving prompts, launching commands, or interacting with privileged windows if an upstream agent is misled or abused.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.