ClawHub
Back to skill

Security audit

Vue

Security checks across malware telemetry and agentic risk

Overview

This is a Vue help skill with disclosed, narrow local preference memory and no executable or network behavior in the inspected artifacts.

Installers should be aware this skill may read and update Vue-specific preference files in ~/Clawic/data/vue/. Avoid storing secrets or sensitive project details there if you do not want them reused in future Vue assistance.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (3)

Description-Behavior Mismatch

Low
Confidence
88% confidence
Finding
The skill instructs persistent storage of user preferences and contextual memory in files under the user's home directory, even though the manifest describes a Vue debugging/review assistant rather than a profile-building system. This expands data retention beyond the immediate task, creating avoidable privacy and data-governance risk if sensitive project details, habits, or policy preferences are written without explicit user awareness and consent.

Context-Inappropriate Capability

Low
Confidence
84% confidence
Finding
The instructions direct the skill to maintain long-lived memory about the user's project shape, libraries, pain points, and explanation preferences, which amounts to ongoing user profiling not clearly required for Vue troubleshooting. Even if intended to improve UX, retaining this context across sessions can accumulate sensitive development patterns and organizational preferences that exceed the principle of data minimization.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The setup instructs the skill to read user-specific files from `~/Clawic/data/vue/` automatically on first use, without any user-facing notice, consent, or contextual disclosure. Silent access to home-directory files is more serious than ordinary app-state handling because those files may contain prior context, preferences, or sensitive notes, and the user is not told that local data will be consulted before responding.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.