CN Content Matrix

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Chinese social-media content generator that searches the web for topic research and saves drafts locally, with no evidence of hidden or destructive behavior.

Install if you are comfortable with topic research being sent to web search and generated content being saved locally. Avoid using confidential campaign plans, customer data, unreleased product names, or sensitive drafts as topics unless you intend them to be searched or written to disk.

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 (1)

Vague Triggers

Medium
Confidence
82% confidence
Finding
Defaulting any unrecognized input to the full content-matrix flow creates an overly permissive trigger that can cause unintended execution of the most expansive workflow. In this skill, that broader path can invoke web searches, read multiple reference files, and write multiple output files, increasing the chance of surprising side effects from ambiguous or malformed user input.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal