Security audit
Interaction Pipeline
Security checks across malware telemetry and agentic risk
Overview
This instruction-only skill changes how the agent reports progress, without code execution, data access, credentials, or persistence beyond the installed instructions.
Install this if you want short, consistent progress updates after actions. Be aware it may conflict with tasks that require strict output formats or with other skills that control response style, but the artifact does not show sensitive access, hidden behavior, or executable code.
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.