Back to skill

Security audit

memory-referee

Security checks across malware telemetry and agentic risk

Overview

This skill appears to be a scoped memory-record cleanup tool that processes provided JSON input and returns reports without hidden credential, network, destructive, or persistent behavior.

Install only if you are comfortable running the npm dependency setup, and provide only the memory records you intend the tool to adjudicate. Review its conflict, stale-record, and classification output before allowing it to change or replace important agent memory downstream.

SkillSpector

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

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.