ClawHub
Back to skill

Security audit

Phonebridge

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly a tutorial, but it handles sensitive phone-data workflows and points to a site whose privacy claims conflict with its tracking script.

Install only if you are comfortable using it as a phone-data tutorial. Before following export steps, confirm you are working on your own device/account, avoid letting an agent read ~/PhoneMirror/ unless you explicitly asked for that, store exported chats securely, and turn off USB debugging after use. Be aware that the linked site currently includes third-party page tracking despite claiming otherwise.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill explicitly helps users export phone data, AI chat history, and WeChat backups, yet the documentation does not warn about privacy, consent, device trust, or handling of exported files. In this context, missing safety guidance increases the risk that users expose sensitive personal or third-party data during backup/export workflows, especially when enabling USB debugging or moving chat records to a PC.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger list includes broad, common phrases such as 'USB调试', '开发者模式', and '手机连电脑', which are likely to appear in ordinary user conversations and can cause unintended skill invocation. In this skill, accidental activation is more concerning because the skill directs users to external web content and potentially local data sources, increasing the chance of unnecessary data exposure or confusing workflow hijacking.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.