ClawHub
Back to skill

Security audit

Airpoint

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Mac-control integration, but it asks users to grant very broad screen and input control without clear safety boundaries or privacy handling.

Install only if you are comfortable giving Airpoint Accessibility and Screen Recording access to your Mac. Avoid using it while secrets, private messages, regulated data, payments, or account settings are visible unless you explicitly intend the agent and configured model providers to process that context. Review Airpoint's app permissions, stored screenshots, and provider API settings before using broad or fire-and-forget tasks.

A.I.G

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

SkillSpector

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)

Missing User Warnings

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill advertises screen reading, screenshot capture, and full Mac control, but it does not provide an explicit user-facing warning about the privacy and security implications of granting Accessibility and Screen Recording permissions. This is dangerous because the agent may access sensitive on-screen data, capture confidential information in screenshots, and make impactful system changes, while users are not clearly warned about the breadth of access or advised on safe usage boundaries.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill exposes a highly privileged computer-use agent through unrestricted natural-language tasking without documenting any guardrails, allowed-task boundaries, confirmation requirements, or prohibited actions. Because the agent can open apps, type, read on-screen content, and manipulate system settings, broad prompting materially increases the risk of unsafe execution, prompt injection via user requests, and unintended destructive or privacy-invasive actions.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.