ppt

• Outcome: Pass
• Latest audit: May 28, 2026, 2:01 AM
• Version:
  1.0.0

Overview

This is a straightforward HTML slide-generation skill, with the main caveat that generated presentations load some assets from public CDNs.

Use this skill for generating HTML presentations from your content. Before opening or sharing the generated HTML, be aware that it may load TailwindCSS and fonts from third-party CDNs; use local or pinned assets for sensitive, offline, or high-assurance presentations.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Description-Behavior Mismatch

Medium

• Confidence: 92% confidence
• Finding: The template loads executable and styling dependencies from third-party CDNs, which means the generated HTML is not truly self-contained and its behavior depends on remote infrastructure at runtime. This introduces supply-chain and privacy risk: a compromised CDN, blocked network, or modified remote asset can change script execution, break rendering, or leak access metadata when the file is opened.

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The trigger condition is broad enough to activate on many ordinary presentation-related requests, which can cause the agent to invoke this skill when the user did not specifically want an HTML slide generator. Overbroad routing increases the chance of unintended tool use, context capture, and incorrect outputs, especially because the skill transforms user content and produces executable HTML.

Vague Triggers

Medium

• Confidence: 87% confidence
• Finding: The default prompt is generic enough to activate broad presentation-generation behavior without clearly constraining scope, inputs, or safety expectations. In an agent setting, overly broad defaults can cause unintended invocation, mismatched outputs, or processing of user content in contexts the skill was not meant to handle.

Natural-Language Policy Violations

Medium

• Confidence: 91% confidence
• Finding: The skill metadata and default prompt hard-code Chinese-language interaction, which can override or conflict with the user's actual language preference. This is mainly a safety and usability issue, but it can also cause misleading outputs, consent issues, and failure to preserve user intent in multilingual environments.

VirusTotal

66/66 vendors flagged this skill as clean.