ClawHub

Philosophy Dialogue

Security checks across malware telemetry and agentic risk

Overview

This philosophy persona skill is mostly a prompt-and-reference workflow; it has broad activation wording and a manual local maintenance script, but I found no automatic execution, network use, credential access, or deceptive behavior.

Install this only if you want philosophy/persona-style responses and are comfortable with broad triggers. Do not run scripts/update_perspective.py unless you intentionally want it to rebuild local philosophy-dialogue reference files from your workspace, replacing existing copied perspective folders.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
74% confidence
Finding
Claiming that any input of a person's name can activate the skill creates an overly broad activation surface that may hijack unrelated conversations. In a meta-skill that impersonates many named figures, this increases the chance of unintended invocation, context confusion, and policy bypass through accidental role-switching.

Vague Triggers

Medium
Confidence
83% confidence
Finding
Generic trigger phrases like '思想家视角' and '多视角讨论' are common conversational language and may collide with ordinary user requests. Because the skill changes response mode and can orchestrate multiple personas, accidental triggering could override intended behavior and produce misleading or unauthorized role-based outputs.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal