NVIDIA LocateAnything-3B vision-language grounding model
Security checks across malware telemetry and agentic risk
Overview
This skill is a documentation-only guide for using NVIDIA LocateAnything-3B and does not include hidden code, persistence, credential use, or automatic actions.
Install only if you are comfortable cloning and running the referenced NVIDIA/SandAI ML repositories. Use a virtual environment or container, review the upstream licenses, and note that the model is described as non-commercial research use.
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.