Back to skill

Security audit

Smart Wake

Security checks across malware telemetry and agentic risk

Overview

This skill is instruction-only and not malware, but it teaches agents to keep working by scheduling wakeups and spawning resume sessions without a clear user approval step.

Install only if you want agents to schedule their own resume wakeups for long-running tasks. Require explicit approval before cron wakeups or spawned resume sessions, avoid sensitive or irreversible work unless the resumed scope is clear, and make sure pending wakeups can be inspected and canceled.

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.