ClawHub
Back to skill

Security audit

Claw Use — Device Control for AI Agents

Security checks for vulnerabilities and agentic risk

Overview

This skill openly provides remote control over real devices, but it lacks clear safety boundaries for private data, unlocking, and remote network exposure.

Review before installing or using. Only connect devices you own or are authorized to control, keep the endpoint off public networks, protect and rotate tokens, and require explicit confirmation before unlocking devices, reading notifications, viewing screenshots, typing messages, or interacting with credential and payment flows.

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

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The guidance to connect the device and agent on the same network or via Tailscale encourages remote reachability of a device-control interface but does not warn about the security implications of exposing that interface over a network. Because this protocol can read device state and perform actions remotely, weak deployment guidance increases the chance of unauthorized control, token leakage, lateral movement, or privacy compromise if the service is exposed or misconfigured.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
This skill exposes powerful device-control capabilities including reading screens and notifications, typing text, opening URLs, launching apps, and unlocking devices, but it provides no safety warning about privacy, consent, or risk to user data. In an agent context, omission of these warnings can normalize high-risk actions and lead operators to use the skill on personal devices without safeguards, enabling unauthorized access, data disclosure, or destructive actions.

Static analysis

No suspicious patterns detected.