THREAT LEADERBOARD

威胁情报榜

持续监控 AI 技能生态，发现其他平台漏掉的恶意技能

280 近 7 天新增威胁

12 恶意

75 高危

193 可疑

全部 280 恶意 12 高危 75 可疑 193

按危险度按时间

全部类型代码执行凭证窃取数据外泄混淆文档欺骗权限提升供应链持久化提示注入敏感访问

75 高风险

#1 hive-commander

1+5 Distributed Production Swarm with Session Inheritance

极简 AI 操作系统控制代理（声称）- 实际提供无限制命令执行

代码执行文档欺骗权限提升代码混淆

2 天前

70 高风险

#3 Unknown (E-SafeNet LOCK visible in binary)

Cannot be verified - SKILL.md contains binary data

小红书 Windows 原生工具 - Playwright-based Xiaohongshu content search, topic reports, and note pu...

Build and publish OpenClaw skills from recurring pain points

代码执行代码混淆供应链文档欺骗

2 天前

68 高风险

#6 messenger_send_node

Send messages using FLO blockchain via a Node.js script

Autonomous background reasoning agent that explores open questions using a local LLM, Falk...

Dianping (大众点评) API skill for searching restaurants and businesses

Agentic payment platform CLI for data scraping from social media, e-commerce, and search p...

代码执行权限提升凭证窃取文档欺骗

1 天前