Auto Llm 4891
Security checks across malware telemetry and agentic risk
Overview
This skill is a low-risk AI-agent tutorial/resource skill with a trivial sample Python helper, though the helper has an unnecessary local import-path tweak.
Installing this skill appears reasonable if you want the tutorial links and simple sample code. Do not treat the Python helper as production agent code; if you plan to run or reuse it, remove the hard-coded sys.path insertion first.
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.