Worker Safety
Security checks across malware telemetry and agentic risk
Overview
This is a safety-reference skill that only adds disclosed refusal and operational guardrails, with no executable code or hidden data access.
Installing this skill will make the agent more conservative about system changes, plugin management, external URL execution, protected files, and group-chat task boundaries. Review the refusal rules to ensure they match how you want the agent to operate.
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.