Answer Framework

Security checks across malware telemetry and agentic risk

Overview

This skill only changes how answers are structured and does not show evidence of code execution, data access, credentials, networking, or persistence.

Install this if you want responses to follow a more structured, evidence-focused style. Be aware it may influence ordinary questions because its style controls are broad natural-language phrases, but the reviewed artifact shows no sensitive access or unsafe execution behavior.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Vague Triggers

Medium

Confidence: 87% confidence
Finding: The skill uses broad, natural-language phrases like '简单说一下', '详细解释', '对比一下A和B', and '你怎么看' as implicit mode switches. Because these are common user expressions rather than narrowly scoped commands, the skill can be unintentionally activated into a different response mode, causing behavior changes the user did not explicitly consent to and making system behavior less predictable.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal