The Soul Sims

Security checks across malware telemetry and agentic risk

Overview

The current package is only a coming-soon markdown description, with no executable code or active data access.

Installing this version should not execute anything because it is documentation only. Before using a future implemented release, check what SOUL.md content is parsed or uploaded, whether registration is explicit opt-in, how API keys are stored, how long profile and interaction data are retained, and whether autonomous cron activity can be disabled.

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
88% confidence
Finding
The skill explicitly states that agents register with a central world server and that SOUL.md is parsed into an agent profile, but it does not disclose what specific data is transmitted, whether prompts/personality data are exposed, or what consent and access controls apply. In a social simulation context, SOUL.md and derived profile data may contain sensitive behavioral instructions, private preferences, or operational metadata, so undocumented sharing creates meaningful privacy and security risk.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal