AI Security AI安全 7h ago Updated 2h ago 更新于 2小时前 52

Over 250 ClickFix Domains Use Browser Fingerprinting to Hide macOS Malware Lures 250多个ClickFix域名使用浏览器指纹技术隐藏macOS恶意软件诱饵

A macOS ClickFix campaign spanning 250+ front-end domains now uses server-side browser fingerprinting to selectively serve malicious lures only to genuine Mac users while hiding content from crawlers, sandboxes, and analysts The ~2.5 KB JavaScript gate probes navigator values, screen dimensions, WebGL signals, timezone, iframe embedding, touch support, developer console activity, and codec support to distinguish real Macs from virtual machines and stealth browsers Qualified victims receive a Git 超过250个前端域名采用浏览器指纹识别技术,服务器端动态决定向访客展示恶意诱饵还是空白页面 攻击门控脚本约2.5KB JavaScript,通过检测平台字符串、屏幕尺寸、WebGL信号、时区、iframe嵌套、触摸支持、开发者控制台计数器和canPlayType调用等10余项指标区分真实Mac用户与爬虫/沙箱 攻击链最终部署AMOS(Atomic Stealer)信息窃取器,窃取凭证、浏览器数据、认证存储、加密货币钱包和敏感文件 攻击仍依赖用户手动复制并运行混淆的Terminal命令,macOS 26.4新增30天未使用Terminal的确认提示和XProtect命令追踪功能 微软建议防御者优

85
Hot 热度
75
Quality 质量
60
Impact 影响力

Analysis 深度分析

TL;DR

  • A macOS ClickFix campaign spanning 250+ front-end domains now uses server-side browser fingerprinting to selectively serve malicious lures only to genuine Mac users while hiding content from crawlers, sandboxes, and analysts
  • The ~2.5 KB JavaScript gate probes navigator values, screen dimensions, WebGL signals, timezone, iframe embedding, touch support, developer console activity, and codec support to distinguish real Macs from virtual machines and stealth browsers
  • Qualified victims receive a GitHub-themed fake download page with a forged "Verified Publisher" badge and are tricked into pasting an obfuscated Terminal command that deploys the AMOS infostealer
  • The fingerprinting gate represents an evasion upgrade against automated analysis rather than a change to the core attack chain, which still relies on social engineering to get users to run Terminal commands
  • Apple's macOS 26.4 introduces a Terminal confirmation prompt for infrequent users and XProtect can trace and block commands pasted from browsers into terminal emulators

Why It Matters

This campaign illustrates the growing sophistication of macOS-targeted social engineering attacks, where threat actors are actively adapting evasion techniques to bypass automated security analysis and sandboxing. For defenders, it highlights the importance of hunting behavioral artifacts like self-submitting fingerprint forms and the mode:"php" indicator rather than chasing disposable front-end domains. The shift from disk image delivery to Terminal-based script fetching represents a broader trend in macOS malware distribution that security teams need to monitor closely.

Technical Details

  • Browser fingerprinting gate: A ~2.5 KB JavaScript script reads navigator.platform (expecting MacIntel), screen/window dimensions, WebGL graphics signals, timezone, iframe embedding status, and touch support. Two analyst-targeting probes include a developer console open-counter and a repurposed canPlayType("video/mp4") call that flags stealth browsers faking codec support. Results are sent server-side with mode:"php" and no user interaction required.
  • Server-side content selection: Requests from crawlers, sandboxes, VMs, or unexpected locations receive blank pages, fake browser extensions, or unrelated business sites. Genuine Mac requests receive a GitHub-themed "Download for macOS" page with a forged "Verified Publisher" badge and ClickFix instructions.
  • Payload chain: Victims who paste and run the obfuscated Terminal command trigger a curl call to a /curl/<id> path, which retrieves additional scripts and launches AMOS (Atomic Stealer), an infostealer targeting credentials, browser data, authentication stores, cryptocurrency wallets, and sensitive files. The broader cluster has also distributed MacSync.
  • Infrastructure: Over 250 front-end domains tracked, many combining "file" with dictionary terms (e.g., filecopperbasket[.]sbs, applefilevault[.]com). Microsoft recommends hunting shared staging infrastructure and /curl/ paths rather than individual disposable domains.
  • Defensive indicators: Monitor for curl piped into zsh, Base64 decoding, osascript, archive creation followed by outbound HTTP POST requests, self-submitting fingerprint forms, hidden fingerprint fields, and the mode:"php" artifact.

Industry Insight

  • Security teams should prioritize behavioral detection of fingerprinting gates and shared staging infrastructure over domain-level blocking, as the disposable nature of front-end domains makes takedown efforts inefficient against this campaign structure.
  • The macOS ecosystem is becoming an increasingly targeted surface for infostealer campaigns adopting evasion techniques previously more common in Windows threat operations; organizations should ensure Terminal usage policies and XProtect signatures are current, especially given macOS 26.4's new confirmation prompt protections.
  • The evolution from disk image delivery to in-browser Terminal command injection reflects a broader shift toward lower-friction, script-based macOS malware distribution that reduces attacker operational overhead while maintaining effective social engineering vectors.

TL;DR

  • 超过250个前端域名采用浏览器指纹识别技术,服务器端动态决定向访客展示恶意诱饵还是空白页面
  • 攻击门控脚本约2.5KB JavaScript,通过检测平台字符串、屏幕尺寸、WebGL信号、时区、iframe嵌套、触摸支持、开发者控制台计数器和canPlayType调用等10余项指标区分真实Mac用户与爬虫/沙箱
  • 攻击链最终部署AMOS(Atomic Stealer)信息窃取器,窃取凭证、浏览器数据、认证存储、加密货币钱包和敏感文件
  • 攻击仍依赖用户手动复制并运行混淆的Terminal命令,macOS 26.4新增30天未使用Terminal的确认提示和XProtect命令追踪功能
  • 微软建议防御者优先狩猎指纹门控特征(自提交表单、隐藏字段、mode:"php"标记)而非追逐一次性前端域名

为什么值得看

该攻击展示了恶意软件分发基础设施从静态HTML向动态指纹门控的演进,反映了网络犯罪团伙对抗自动化分析的技术升级。对安全从业者而言,理解这种"差异化响应"机制有助于调整威胁狩猎策略,从域名黑名单转向行为特征检测。

技术解析

  • 指纹识别门控机制:约2.5KB JavaScript脚本读取navigator.platform(验证MacIntel)、屏幕/窗口尺寸、WebGL图形信号,结合时区、iframe嵌套检测、触摸支持(桌面Mac通常无)等多维特征,区分真实Apple硬件与虚拟机/模拟器环境。
  • 分析师对抗探针:两项专门针对安全研究的检测——开发者控制台打开计数器(每次检测递增)和canPlayType("video/mp4")调用作为陷阱,识别伪装编解码器支持的隐身浏览器。
  • 服务器端动态响应:指纹数据以mode:"php"标记无交互提交至服务器,服务器为爬虫/沙箱返回空白页、虚假扩展或无关商业网站,为符合条件的Mac请求展示带伪造"Verified Publisher"徽章的GitHub主题下载页。
  • 攻击链架构:用户运行混淆命令后访问/curl/路径获取后续脚本,最终部署AMOS信息窃取器;同一基础设施集群还分发MacSync,微软未映射所有域名与载荷的对应关系。
  • 检测指标建议:监控浏览器访问后异常Terminal活动,特别是curl管道至zsh、Base64解码、osascript调用、归档创建及出站HTTP POST请求;狩猎指纹表单自提交、隐藏字段和mode:"php" artifact。

行业启示

  • 威胁基础设施演进趋势:恶意软件分发正从静态诱饵页转向动态指纹门控,攻击者通过差异化响应规避自动化分析,安全防御需从域名封锁转向行为特征和通信模式检测。
  • macOS安全生态变化:Apple macOS 26.4引入Terminal确认提示和XProtect命令追踪,反映平台安全重心从被动防护转向主动用户教育和命令级监控,企业应同步更新终端安全策略。
  • 防御策略调整建议:面对一次性域名和共享基础设施并存的攻击模式,组织应优先建立对指纹门控特征的检测能力,结合网络层/curl路径封锁和终端行为监控,而非依赖传统的域名黑名单维护。

Disclaimer: The above content is generated by AI and is for reference only. 免责声明:以上内容由 AI 生成,仅供参考。

Security 安全 Research 科学研究