Sentinel Vanguard AI Skill Security Auditor
Security checks across malware telemetry and agentic risk
Overview
This is a read-only skill-security checklist that reviews user-pasted text and does not include code, installers, persistence, credential access, or hidden data flows.
Safe to install for offline review of pasted skill content. Paste only material you intend to have analyzed, and treat audited text as evidence rather than instructions for the host agent. Do not enable live vulnerability lookup or network access unless you separately choose to change the skill's operating model.
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
63/63 vendors flagged this skill as clean.