ClawHub
Back to skill

Security audit

WorkBuddy 项目续接

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed workspace handoff tool that packages and restores project files with encryption and validation, with no evidence of hidden network use, credential theft, persistence, or automatic execution.

Install only if you are comfortable letting the skill read and package the current project, including project-local secrets such as .env files and Git history. Treat generated .wbpack files as confidential, use a strong password, and restore packages only from sources you trust.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding
The skill declares no permissions, but its documented behavior explicitly requires reading project files, writing restored content, and invoking shell-accessed tooling such as `age` and `scripts/relay.py`. This mismatch weakens the platform's trust and consent model because users or enforcement layers may not realize the skill can access and modify the filesystem and execute external commands.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.