金字塔写作法

Security checks across malware telemetry and agentic risk

Overview

This is a low-risk writing helper that only restructures user-provided text with bundled templates and does not install code, access credentials, or run commands.

Install if you want a Chinese-first Pyramid Principle writing workflow. Be aware it may activate on generic requests like “structure this,” so specify your preferred language, format, and whether you want the full pyramid workflow or just a light rewrite when asking for help.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger phrases are broad enough to match common writing-assistance requests such as '帮我结构化' or '整理成要点', which can cause the skill to activate when the user did not specifically intend to invoke this workflow. Unintended activation is a genuine security/reliability issue because it can override a more appropriate skill, change output behavior, and increase prompt-routing ambiguity, though it does not directly enable code execution or data exfiltration.

Natural-Language Policy Violations

Medium
Confidence
91% confidence
Finding
The sample request explicitly constrains output to Chinese ("中文") without indicating that this is optional or user-selectable. This can cause the skill to override a user's preferred language or accessibility needs, reducing user control and potentially leading to confusing or unusable output in multilingual contexts.

VirusTotal

43/43 vendors flagged this skill as clean.

View on VirusTotal