ClawHub
Back to skill

Security audit

ppt-prompt-distiller

Security checks across malware telemetry and agentic risk

Overview

This skill is a prompt-writing helper that stays within its stated purpose, with the main caveat that some Chinese trigger phrases are broad.

Install this if you want a Chinese-language PPT prompt distillation workflow. Be aware that short phrases like “蒸馏” may invoke it more easily than a narrowly named trigger, so use explicit wording when you do or do not want this skill involved.

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 (4)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger list includes generic phrases such as '蒸馏' and '转提示词' that can plausibly appear in ordinary user requests unrelated to this specific skill. In a skill-based agent, overly broad activation terms can cause unintended invocation, leading the agent to process content under this skill's workflow when the user did not explicitly intend that behavior.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger phrases are broad functional terms such as '蒸馏', '转提示词', and '生成PPT提示词', which can match many ordinary user requests unrelated to this specific skill. Over-broad activation can cause the wrong skill to take control, leading to unintended prompt transformations, confusion, and possible interference with safer or more appropriate agent behaviors.

Natural-Language Policy Violations

Medium
Confidence
84% confidence
Finding
The skill description is written in Chinese and strongly implies Chinese-language operation without stating locale limits or offering language fallback. This can cause incorrect activation or degraded output quality for users operating in other languages, and may make behavior less predictable in multilingual environments.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The phrase "看着办" is highly generic conversational language, so it can unintentionally trigger the skill in unrelated contexts. In an agent environment, ambiguous invocation broadens the attack surface and can cause the skill to activate on loosely related user input, leading to unintended prompt transformation or workflow execution.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.