Skill Design Guide
Security checks across malware telemetry and agentic risk
Overview
This is an instruction-only skill design guide that does not request sensitive access, execute code, or persist changes.
This skill appears safe to install as a guide for skill design and review. As with any low-trust publisher package, review future updates before installing if they add scripts, hooks, credential use, network calls, or automatic changes to agent behavior.
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.