ClawHub

把5000年华夏文明压成24个可调用的思维框架。用老祖宗的思维帮你破局，覆盖决策、效率、成长、关系、搞钱、生活6大场景，每个框架有独特的思考工具和输出结构。AI自动匹配用户状态，也支持手动指定和跨模块组合。半文半白的"老祖宗口吻"输出，像古代圣贤坐在你对面聊天。

Security checks across malware telemetry and agentic risk

Overview

This is a text-only coaching skill with broad automatic triggers, but it does not request code execution, credentials, data access, persistence, or account-changing authority.

Install this if you want a stylized, opinionated advice framework. Consider invoking it explicitly if you do not want ordinary conversations routed into this style, and independently verify any high-stakes legal, medical, financial, or career advice before acting.

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
94% confidence
Finding
The skill description defines activation using very broad psychological and conversational states like纠结、焦虑、竞争压力 and also says the skill should主动推荐 for many deep-thinking questions. This can cause unintended invocation during ordinary conversations, leading the agent to steer users into this skill when they did not explicitly request it, which is a scope-control and reliability problem.

Vague Triggers

Medium
Confidence
96% confidence
Finding
The default auto-matching matrix maps many common phrases and states to framework invocation without clear thresholds or disambiguation. Because these cues are ubiquitous in normal conversation, the system may over-trigger, override user intent, or route sensitive interpersonal, career, or mental-stress discussions into a preset advisory style without sufficient consent or contextual checks.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal