ClawHub
Back to skill

Security audit

LYGO Champion Council (Δ9 unified v2)

Security checks across malware telemetry and agentic risk

Overview

This is a low-risk advisory roster skill; it may overstate persona-selection functionality, but the included scripts only read local reference files and it explicitly limits privileged actions to user consent.

Install this if you want a lightweight LYGO council roster and advisory prompt reference. Do not expect it to enforce persona routing or perform stack operations by itself, and keep following the skill's own guidance: do not share API keys or .env contents, and require explicit consent before any shell, publishing, vault, or account-affecting action.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding
Without declared permissions the skill's intent is opaque and cannot be validated.

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding
This is a mismatch because the description presents the skill as a functional multi-persona selector/install mechanism for 15 champions, including selection by champion_id or egg_id and legacy slug compatibility. The actual code does not implement selection, persona routing, advisor-only behavior, or legacy compatibility behavior. Instead, it provides two utility scripts: one to print roster entries and one to validate local files and roster count. The primary purpose of the code is inspection/self-checking of metadata, which differs materially from the declared functionality.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.