Self-Governor
Security checks across malware telemetry and agentic risk
Overview
This no-code decision skill is transparent, but it can steer an agent to keep going around costly or irreversible actions without requiring user confirmation.
Install only if you want the agent to make lightweight autonomous next-step decisions. Before using it with tools that can publish, delete, overwrite, modify accounts, or spend credits, enforce a separate rule that those actions require explicit user confirmation.
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
50/50 vendors flagged this skill as clean.