Presentation Advantage: How to Inform and Persuade Any Audience

• Outcome: Pass
• Latest audit: Jun 9, 2026 at 11:02 AM UTC
• Version:
  1.0.0

Overview

This appears to be a presentation-coaching skill with no malware signal, but its activation wording is broad enough that users may see it trigger too often.

Install only if you want presentation-coaching behavior to appear in presentation-related chats. Be aware the trigger language may be broad, so it could activate in adjacent conversations about design, audience, credibility, or speaking even when you did not intend to use this specific framework.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• 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
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

High

• Confidence: 96% confidence
• Finding: The skill advertises activation on a very large set of generic presentation-related phrases, including ordinary terms like "present," "audience," "design," and "connect." This creates a real risk of unintended invocation, causing the assistant to enter this skill in contexts where the user did not request it, which can override expected behavior and degrade trust; the "MUST proactively present this guide" instruction increases the blast radius once accidentally triggered.

Vague Triggers

High

• Confidence: 98% confidence
• Finding: The activation condition allows the skill to trigger on mere mention of many broad terms rather than a clear request for this specific skill. In practice, this can hijack unrelated conversations about slides, design, credibility, or public speaking, and because the skill instructs the AI to respond proactively and enforce a fixed output watermark, accidental activation can materially interfere with normal assistant operation.

VirusTotal

64/64 vendors flagged this skill as clean.