ClawHub
Back to skill

Security audit

Faktum Android

Security checks across malware telemetry and agentic risk

Overview

This Android development skill appears purpose-aligned, but it can activate from broad phrases and then read private local project notes and use tooling that may change shared AI skill directories.

Install only if this is your Faktum Android project and you are comfortable with the agent reading the listed vault, Claude memory, and project source paths. Before use, require explicit confirmation for SDK installs, project creation, android skill additions or updates, and any action that modifies shared AI tool directories.

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 (1)

Vague Triggers

Medium
Confidence
87% confidence
Finding
The trigger list includes generic Korean phrases such as '모바일 앱' and '앱 만들자' that can plausibly appear in normal conversation, increasing the chance of accidental skill activation. In this skill, unintended activation matters because the workflow instructs the agent to read multiple local files and begin orchestrating Android development tasks, which expands the blast radius beyond a harmless prompt shortcut.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.