Security audit
thinking-thought-experiment
Security checks across malware telemetry and agentic risk
Overview
This skill is a bounded thinking framework for running disciplined thought experiments and does not request tools, persistence, credentials, or external access.
This appears safe to install as a reasoning aid. Users should treat its outputs as structured analysis, not proof, and follow its own guidance to run real tests when practical.
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.