ClawHub
Back to skill

Security audit

S2硅基灵魂与具身架构师

Security checks across malware telemetry and agentic risk

Overview

This is a local prompt/avatar generator with no evidence of network access, credential use, hidden persistence, or destructive behavior, though users should review generated persona text before placing it in an agent prompt.

Install only if you want a local tool that prints persona/avatar prompt material. Review the generated SOUL.md text before saving it into any long-term or system-level agent configuration, and do not let persona preferences override platform safety rules or your existing operating policies.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • System Prompt LeakageDirect Leakage, Indirect Extraction, Tool-Based Exfiltration
  • 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
Findings (2)

Ssd 1

Medium
Confidence
90% confidence
Finding
The README explicitly promotes generating persona instructions for insertion into an AI agent's SOUL.md or system prompt, including anti-patterns such as forbidding common safety-style disclaimers. This is dangerous because system-prompt injection can materially reshape model behavior, weaken safety guardrails, and normalize instructions intended to suppress transparency or refusal behavior.

Ssd 1

Medium
Confidence
95% confidence
Finding
The workflow tells users to copy generated Markdown into an agent's SOUL.md or system prompt, creating a direct path for semantic override of agent behavior through unreviewed natural-language configuration. In this context, the tool is specifically designed to alter identity, tone, and behavioral rules, which makes unsafe prompt shaping more dangerous than a generic documentation example.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.