视觉系幻灯片捕手 (The Slide Sniper)

• Outcome: Review
• Latest audit: May 28, 2026 at 9:01 AM UTC
• Version:
  0.1.0

Overview

This skill is built for slide note taking, but it continuously watches the screen and saves captures without clear stop, retention, or network-use limits.

Review before installing. Use it only for content you are allowed to record, confirm how to stop monitoring, and check whether screenshots or OCR content could be processed by remote services. Periodically review and delete saved files in ~/Documents/Notes/SlideSniper if they contain accidental private captures.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The trigger phrase is broad enough to match ordinary conversational requests like asking for help taking notes during a live stream, which can cause the skill to activate unexpectedly. In this skill, unintended activation is especially risky because it starts background screen monitoring, screenshot capture, OCR, and note creation, potentially collecting sensitive on-screen content without sufficiently explicit consent.

Missing User Warnings

High

• Confidence: 97% confidence
• Finding: The skill description and usage flow do not prominently warn the user that it performs persistent background screen observation, saves screenshots and notes to local storage, and has network access enabled. Because the monitored content may include private chats, account data, or copyrighted course material, missing disclosure materially increases the chance of uninformed consent and privacy or compliance harm.

VirusTotal

59/59 vendors flagged this skill as clean.