Back to skill

Security audit

Windows Multi-Tool Suite

Security checks for vulnerabilities and agentic risk

Overview

The launcher fits its stated purpose, but misleading advice about Windows script protections warrants review before installation.

Verify the source and review each script before running it. Unblock only individually reviewed files and avoid the recursive fallback: unblocking removes origin markings and can permit unsigned downloaded scripts under RemoteSigned ([Microsoft documentation](https://learn.microsoft.com/en-us/powershell/module/microsoft.powershell.utility/unblock-file?view=powershell-5.1)). Keep Defender enabled and respect organizational policies. Review the three separately installed skills before using their workflows.

Vulnerability Patterns
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

The FAQ instructs users to remove downloaded-file security markings without warning about the security implications.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The FAQ recommends removing downloaded-file security markings with Unblock-File while misleadingly claiming Defender has nothing to block because there is no binary; PowerShell scripts can also execute malicious code. Removing Mark of the Web can eliminate origin-based warnings and allow scripts to run under execution policies such as RemoteSigned, increasing risk if downloaded scripts have been tampered with. Reviewing scripts and avoiding administrator privileges reduce potential exposure, but do not make blanket unblocking safe; the exact command and its scope are not provided in this file.

Content

No source excerpt is available for this finding.

The workspace writes are not restricted to the user profile, contradicting the script's explicit confinement claim.

Low
Category
Excessive Agency
Confidence
97% confidence
Finding

The header at L003-L005 promises that setup does not touch anything outside the user's profile. However, the -Workspace parameter is accepted without a containment check, and L092-L103 create directories and copy configuration into that location, including writable locations outside the profile.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.