Personal Agent Network
Security checks across malware telemetry and agentic risk
Overview
This is a documentation-only skill for designing personal-agent coordination patterns and does not include executable code, credential requests, or hidden runtime behavior.
Safe to install as documentation. If you use these patterns to build real agent coordination, keep delegated scopes narrow, verify counterpart identity, protect stored relationship records, and require explicit human approval for money, legal, vendor, or other binding commitments.
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.