ClawHub

PPT 生成器 AI Pro

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed PPT-generation integration that uses AiPPT credentials and uploads user-selected content to AiPPT to create and download presentations.

Install only if you are comfortable sending prompts, uploaded documents, and webpage-derived content to AiPPT's service. Use scoped AiPPT API keys, avoid confidential files unless your AiPPT account and policies allow it, and clear any local token cache if you rotate or remove credentials.

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 (3)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The skill description is broad enough to trigger on many ordinary presentation-related requests, which can cause the agent to invoke this skill outside a narrowly intended scope. Over-broad activation increases the chance of inappropriate tool selection, user confusion, and unintended handling of requests that should remain in general chat or be routed elsewhere.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The usage examples use generic phrases like making a product launch PPT or investor deck without defining boundaries, making them likely to match a wide range of normal user requests. This can bias the routing system toward this skill too aggressively and lead to unintended invocation for loosely related presentation assistance.

Natural-Language Policy Violations

Medium
Confidence
83% confidence
Finding
The skill content is entirely Chinese-oriented and emphasizes Chinese typesetting without offering a language-selection path, which can force operation in one language regardless of user preference. This is mainly a safety/quality issue because it can produce incorrect-language outputs, confuse users, or mishandle multilingual requests, but it is not directly a code-execution or data-exfiltration risk.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal