get QA pairs from text or session

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:37 PM UTC
• Version:
  1.0.0

Overview

This is an instruction-only skill for turning chats or documents into Q&A and preference summaries, with a real privacy consideration around full-chat extraction but no hidden execution or exfiltration behavior.

Install this only if you are comfortable with the agent summarizing full conversations or uploaded documents into reusable Q&A and preference records. For private chats, ask it to process a specific excerpt or document and review the output before saving or sharing it.

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 (4)

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The trigger description is very broad and matches common requests like summarizing a conversation, extracting decisions, or making flashcards. That can cause the skill to activate in situations where the user did not clearly intend a full conversation-wide extraction, increasing the chance of over-collection and disclosure of prior chat content.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill instructs extraction of inferred preferences from the full conversation but does not warn the user that it may analyze earlier messages and derive profile-like information from them. This undermines informed consent and can surprise users by surfacing sensitive preferences or decisions they did not expect to be compiled.

Ssd 3

Medium

• Confidence: 96% confidence
• Finding: The skill is designed to scan the current chat session and transform it into reusable Q&A and preference summaries, which creates a direct path for broad disclosure of previously shared user information. Because it emphasizes extracting 'every' exchange and implied preferences, it may reveal sensitive content that is not necessary for the immediate request.

Ssd 3

Medium

• Confidence: 97% confidence
• Finding: The workflow and tips explicitly direct scanning from the first message and compiling all Q&A exchanges and inferred preferences, encouraging full-session retention and re-exposure in a structured output. In context, this is more dangerous because the skill's purpose is to convert conversational history into a reusable knowledge base, increasing persistence and visibility of potentially sensitive user data.

VirusTotal

64/64 vendors flagged this skill as clean.