ClawHub
Back to skill

Security audit

LYGO Deception Radar

Security checks across malware telemetry and agentic risk

Overview

This skill is a narrowly scoped public-sample dashboard builder with disclosed local file reads/writes and no evidence of hidden network, subprocess, credential, or destructive behavior.

Install only if you intend to run a local public-sample dashboard builder. Keep LYGO_STACK_ROOT pointed at a trusted LYGO checkout, review any external labeled suite before using it, and do not use the output as a verdict about real people or private communications.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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)

Context-Inappropriate Capability

Low
Confidence
84% confidence
Finding
The script searches for a Python module in multiple external filesystem locations and in a path derived from the LYGO_STACK_ROOT environment variable, then prepends that directory to sys.path and imports lygo_ops_detector. If an attacker can place a malicious file at one of those probed locations or influence the environment, they can cause arbitrary code execution when this builder runs.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.