ClawHub
Back to skill

Security audit

mouse-keyboard

Security checks for vulnerabilities and agentic risk

Overview

This skill appears intended for real Mac mouse and keyboard automation, but it gives broad live-desktop control without clear safeguards and also suggests AppleScript automation.

Install only if you want an agent to control your live macOS desktop. Before use, keep sensitive apps out of focus, verify coordinates and target windows, supervise actions that could submit forms or change data, and avoid AppleScript unless you intentionally allow broader Mac automation.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This skill can generate real mouse and keyboard input against whatever application currently has focus, which can trigger unintended clicks, form submissions, message sending, or destructive actions. Without an upfront warning about active-window interaction and command execution side effects, users may invoke it in unsafe contexts and cause security-relevant or irreversible actions.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill is presented as a cliclick-based mouse/keyboard automation capability, but the note recommending osascript expands it into broader AppleScript-driven automation. That materially increases capability and risk because osascript can drive applications, dialogs, and system actions beyond simple pointer/keystroke injection, while remaining undocumented in the skill scope and safeguards.

Content

No source excerpt is available for this finding.