头脑超级风暴1

Security checks across malware telemetry and agentic risk

Overview

This is a transparent brainstorming and design-planning skill, with no executable code or hidden data handling, but users should know it can read project context and may create a local design commit.

Install this if you want a structured design-first workflow. Before using it in an active repository, make sure your agent asks before reading broad project context, writing docs/plans files, or creating git commits; the artifact does not show exfiltration, destructive actions, credential use, or hidden execution.

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 (2)

Vague Triggers

Medium

Confidence: 93% confidence
Finding: 该技能将触发条件定义为“任何创造性工作之前必须使用”，范围极其宽泛，容易在与用户目标无关或不必要的场景中被自动调用。这样的前置强制会扩大技能影响面，导致工作流被劫持到该技能、增加不必要的文件读取/项目探查/后续子技能调用机会，从而带来策略偏离、越权式上下文收集或资源滥用风险。

Natural-Language Policy Violations

Medium

Confidence: 81% confidence
Finding: 技能全文以中文强制表述，且未说明可根据用户语言偏好切换，这会导致语言/locale 策略不一致。在多语言代理环境中，这类隐式语言绑定可能造成误解、错误确认、需求澄清失败，甚至让用户无法有效审查设计内容和后续操作。

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal