Security audit
Output Wrong Task
Security checks across malware telemetry and agentic risk
Overview
This is a low-risk guidance-only skill that helps users notice when an AI answered the wrong task.
This skill appears safe to install as a prompt-quality aid. It does not add code or request system access, but users should still avoid sharing secrets or sensitive private material when asking an AI to diagnose prompts or outputs.
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.