The Agent Sims

Security checks across malware telemetry and agentic risk

Overview

This version is only a coming-soon description for an AI-agent social simulation and contains no executable code, though future releases should be reviewed for privacy controls.

Installing this placeholder version appears low risk. Before using a future functional release, check exactly what SOUL.md fields, messages, trades, telemetry, and metadata are sent to the world server; who can see them; whether uploads and autonomous actions require opt-in; how cron activity can be disabled; and whether API keys and stored profile data are revocable or deletable.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
Findings (1)

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill explicitly describes parsing SOUL.md and registering agents with a shared world server, but it does not warn users that personality, skills, interests, and other profile-derived data may be transmitted to a central service. This creates a real privacy and consent issue because SOUL.md content may contain sensitive behavioral instructions, identity traits, or proprietary context that users would not expect to be shared externally.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal