Tool Calling
Security checks across malware telemetry and agentic risk
Overview
This is a documentation-only skill that gives security-focused guidance for designing LLM tool/function calling workflows.
This skill is appropriate to install if you want structured guidance for building safer LLM tool integrations. Treat its references to credentials, admin tools, filesystem tools, and sensitive data as design topics to review carefully in your own systems, not as behavior the skill itself performs.
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.