Guard
Security checks across malware telemetry and agentic risk
Overview
This is a guidance-only AI guardrails skill with no code execution, install hooks, credential use, or hidden privileged behavior.
Safe to install as a guidance-only skill. If you use it to build a real production guardrails system, be deliberate about what gets logged, who can review logs, how sensitive content is redacted, and how long records are retained.
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
65/65 vendors flagged this skill as clean.