Security audit
thinking-model-combination
Security checks across malware telemetry and agentic risk
Overview
This skill is a self-contained reasoning guide and does not request file, network, credential, persistence, or execution authority.
This skill appears safe to install as a reasoning aid. It may make responses more structured for complex decisions, but it does not itself grant new system access or perform actions.
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
64/64 vendors flagged this skill as clean.
Static analysis
No suspicious patterns detected.