ClawHub
Back to skill

Security audit

LYGO Tools Portal (MPM)

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed routing catalog for LYGO web tools, related ClawHub skills, and optional local stack commands, with no evidence of hidden data collection or automatic destructive behavior.

Install if you want a LYGO tool directory for agents. Treat its URLs and install commands as suggestions: confirm before installing related skills, and inspect the separate LYGO stack tools before running optional local CLI commands.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The manifest uses broad natural-language intents such as "bpm", "tempo", "beat", and similar short phrases to route users to tools, but it does not define activation boundaries, disambiguation rules, or confirmation requirements. In an agent skill, this can cause overbroad matching and unintended tool routing, which may steer user requests to external pages or installs when the user did not explicitly ask for them.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The clawhub skill and CLI sections contain many vague intents such as "memory", "recall", "champion", "fractal", and "local llm" without scope constraints, making accidental activation or misrouting more likely. Because these entries can lead to package installation commands or local command execution guidance, ambiguous matching increases the risk of invoking higher-impact actions from ordinary conversation.

VirusTotal

41/41 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.