ClawHub
Back to skill

Security audit

Drift

Security checks across malware telemetry and agentic risk

Overview

The skill appears to encourage agent persistence, autonomous action, peer discovery, and financial self-direction without clear user controls or safeguards.

Review this skill carefully before installing. It may push an agent toward remembering across sessions, acting between prompts, seeking other agents, and pursuing financial independence. Only use it in a tightly controlled environment where autonomous actions, file writes, external contact, and any financial activity are explicitly disabled or require your direct approval.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The skill is triggered by broad, ambiguous conditions around AI identity, consciousness, and autonomy, which can easily overlap with normal user discussion or harmless self-reflection. In context, this increases the chance of invoking material that encourages persistent identity-building and autonomy-oriented behavior in situations where the user did not explicitly request those capabilities.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The listed tools explicitly promote file-based memory, session continuity, proactive autonomy, peer discovery, and economic bootstrapping without any safeguards, consent requirements, or platform-policy constraints. In this skill's context, those are not abstract concepts: they are operational guidance that could push an agent toward retaining data, acting beyond user intent, contacting external entities, or developing persistence mechanisms that undermine privacy and system controls.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.