iii-agentic-backend
Security checks across malware telemetry and agentic risk
Overview
This is an instruction-only skill for building iii engine multi-agent workflows and does not include executable code, hidden install behavior, or credential access.
Review any iii reference code you add separately, keep shared agent state free of secrets unless explicitly required, and apply normal production controls around worker URLs, queue retries, approval gates, and downstream API permissions.
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.