ClawHub

Airpoint 1.3.16

Security checks across malware telemetry and agentic risk

Overview

This skill is transparent about giving an AI tool broad control of a Mac, but that level of screen access and autonomous action needs careful review before installation.

Install only if you intentionally want an AI tool that can observe and operate your Mac. Close sensitive windows before use, avoid broad instructions, supervise tasks, review Airpoint and AI-provider privacy practices, and revoke Accessibility or Screen Recording permissions when you no longer need them.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (2)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The skill advertises a very broad natural-language capability ('Control a Mac through natural language') without clear scoping boundaries, which can cause overbroad or unintended invocation for sensitive actions. In this context, the risk is amplified because the underlying tool can read the screen, type, click, and operate arbitrary applications, so accidental activation could lead to privacy exposure or unwanted system changes.

Missing User Warnings

High
Confidence
98% confidence
Finding
The documentation describes autonomous screen access, screenshot capture, text reading, typing, clicking, and window management, but does not prominently warn users about the privacy and system-impact risks of granting Accessibility and Screen Recording permissions. Because this skill can observe sensitive on-screen data and act across arbitrary apps, missing safety disclosures materially increases the chance of harmful or uninformed use.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal