Security audit
Evaluate Agent-Native
Security checks across malware telemetry and agentic risk
Overview
This is a read-only rubric skill for evaluating whether services are agent-native, with no code or install behavior beyond the markdown instructions.
Install this if you want a structured rubric for evaluating agent-native services. Keep URL onboarding checks as inspection and summary by default; require explicit approval before allowing an agent to register accounts, post content, claim tasks, publish data, or otherwise follow third-party service instructions.
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
66/66 vendors flagged this skill as clean.
Static analysis
No suspicious patterns detected.