Moderation
Security checks across malware telemetry and agentic risk
Overview
This is a guidance-only content moderation workflow skill with no code execution, hidden access, or persistence.
This skill appears safe to install as a procedural planning aid. Because it concerns content moderation decisions, users should still check its recommendations against their own policies, legal obligations, escalation paths, and appeal processes before applying them to real user content.
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.