ClawHub
Back to skill

Security audit

Magneto Skill Master

Security checks across malware telemetry and agentic risk

Overview

This skill is transparently meant to install external skills, but it can download third-party code, overwrite local skill folders, install dependencies, and even suggests disabling sandboxing without a clear confirmation gate for every system-changing step.

Review each target repository yourself before using this skill. Require explicit confirmation before cloning, replacing an existing skill, installing dependencies, disabling sandboxing, or running validation commands, and keep backups of any existing ~/.workbuddy/skills/<name>/ directory.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

High
Confidence
94% confidence
Finding
The trigger definition is extremely broad and covers common requests like downloading or installing a skill from external repositories. This can cause the skill to activate in many contexts where the user may not have intended to delegate repository retrieval, auditing, and installation actions, increasing the chance of unintended code acquisition or local modification.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The activation condition includes essentially any scenario involving installing an external skill locally, which is too ambiguous for a capability that downloads code, modifies files, and installs dependencies. In context, this skill is more dangerous because its workflow explicitly clones repositories, copies files into the user skill directory, and may install Python packages, so accidental invocation has real system impact.

Vague Triggers

Medium
Confidence
93% confidence
Finding
The trigger phrases are broad, generic, and overlap with common user requests such as installing or adding a skill. This can cause unintended activation of a skill that performs high-impact actions like downloading, auditing, and installing external content, increasing the chance of unsafe automation without explicit user intent.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The manifest description states that the skill can download external skills, install dependencies, and modify the local WorkBuddy skill directory, but it does not clearly warn users about these system-changing actions. In this context, the omission is more dangerous because the skill's core purpose is handling untrusted external packages and making local changes, which can lead to code execution or persistence if users are not adequately informed.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.