ClawHub
Back to skill

Security audit

Bohrium Sandbox (`lbg sdbx`)

Security checks across malware telemetry and agentic risk

Overview

The skill appears to describe user-directed sandbox workflows, with a real but disclosed storage-mount risk users should handle carefully.

Install only if you understand that mounting user or share storage gives sandboxed commands access to those files. Use mounts only for the specific folders needed, avoid mounting sensitive data when running untrusted uploaded code, and review commands before allowing writes or network access.

SkillSpector

By NVIDIA
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 (1)

Missing User Warnings

Medium
Confidence
86% confidence
Finding
The skill explicitly recommends `--mount-user-storage` and mounting share disks, but it does not warn that doing so exposes potentially sensitive user/project data inside a sandbox where arbitrary shell commands are then executed. In this context, the danger is increased because the same document encourages running uploaded code and interactive shell commands, so mounted storage can become an easy path to unintended data access, modification, or exfiltration.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.