Ai Agent Builder
Security checks across malware telemetry and agentic risk
Overview
This is an instruction-only AI-agent builder skill; its powerful examples are disclosed and aligned with the stated purpose.
Before installing, review any generated agent configuration carefully, connect only tools the agent truly needs, prefer read-only or least-privilege credentials, avoid storing secrets in memory, and monitor any deployed agent that can send email, query databases, collect data, or continue running after launch.
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.