ClawHub
Back to skill

Security audit

iBook Widget Packer

Security checks across malware telemetry and agentic risk

Overview

The skill does the advertised iBook widget packaging work, but its helper script can recursively delete an unintended local folder if given an unsafe widget name or output path.

Install only if you are comfortable with a local packaging helper that reads your project and writes output files. Use a fresh output directory and a simple widget name containing only letters, numbers, hyphens, or underscores. Do not use slashes, absolute paths, or '..' in the widget name, and check the target path before running because an existing matching .wdgt folder may be deleted and recreated.

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 (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill clearly instructs reading source project files and writing multiple output artifacts, but it declares no permissions or trust boundary information. This creates a capability-transparency problem: users and the hosting platform may not realize the skill can access local files and generate archives, increasing the risk of unintended file access or overwrite during execution.

Missing User Warnings

Low
Confidence
88% confidence
Finding
The skill performs file creation, copying, modification, and packaging operations, but it does not prominently warn users that it will generate and rewrite files in the output path. In practice, unclear disclosure can lead to accidental overwrites, packaging of unintended content, or confusion about where modified assets are being written.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal