每日安全资讯（2026-10-07）

• SecWiki News
  • SecWiki News 2026-10-06 Review
• Light Cube
  • LightCube 11 周年：AI 一天，人间一年
• Private Feed for M09Ic
  • bolucat released 202610070033 at bolucat/Archive
  • esrrhs contributed to esrrhs/spp
  • strands-agents released python/v1.58.1 at strands-agents/harness-sdk
  • anthropics released v2.1.292 at anthropics/claude-code
  • timwhitez starred microsoft/nvx
  • Mr-xn starred Niko1221/Strata
  • b1nhack starred google/kernel-research
  • wuhan005 contributed to sayrud/sayrud
  • anthropics released v2.1.291 at anthropics/claude-code
  • pydantic released clai2-bleeding at pydantic/pydantic-ai
  • bolucat released 202610060156 at bolucat/Archive
• Tenable Blog
  • How to mitigate the risk from AI-generated apps built by your 'citizen coder' employees
• Armin Ronacher's Thoughts and Writings
  • What is Codemode
• Sploitus.com Exploits RSS Feed
  • Exploit for Incorrect Authorization in Qualcomm Aqt1000_Firmware
  • bbot exploit
  • CVE-2019-15107 exploit
  • h2conn-exploit
  • CVE-2023-41892 exploit
  • gitGRAB exploit
  • AI-Vulnerabilities-Playground exploit
  • pwnage24mtk exploit
  • CVE-2021-21985_PoC exploit
  • hollows_hunter exploit
  • stackrox exploit
  • cve-2024-3400-poc exploit
  • CVE-2025-30208 exploit
  • exfilkit exploit
  • Research_Successful_Errors exploit
  • Freeloader exploit
  • TrashEmail exploit
  • CredsLeaker exploit
  • BlackWidow exploit
  • BlueSpy exploit
• Doonsec's feed
  • Sub2api 存在支付回调计费漏洞 (中转站长注意)
  • 大厂安全老哥卖“AI渗透Skill技能包”，赚8万，自己却进局子了
  • 德国前联邦情报局局长涉嫌叛国性间谍活动被捕
  • 【入门实操】：手把手教你如何利用智能体快速上手研判
  • 杀人祭天，立投名状！缅北诈骗犯陈大卫啥来头
  • 十月六日，归钓记
  • 国家密码管理局征集国家密码科学基金第三批项目指南建议
  • 自主漏洞发现与利用生成
  • 通过url跳转拿下赏金
  • 被裁第47天，我用AI挖漏洞赚到了第一笔赏金
  • 看我如何使用Trae全程用AI开发了Dota游戏
  • 风采
  • 这3个地方，治愈你的精神内耗
  • 2026 安徽 AI 大模型竞赛 · 网络赛 · 本科软件组 A卷
  • 赵长鹏罕见受访：真实身家100-300亿美元，2013年大胆卖房90万美元全仓比特币
  • iOS 27.2 测试版推进到 Beta 3：这轮更新到底改了什么
  • 德国前情报局长被捕，近2000份内部文件牵出机密交易指控
  • 俄罗斯实验室员工死亡与肺鼠疫泄漏疑云
  • 焊上 JTAG 调试硬盘：西数 WD 固件从压缩格式拆到热补丁延迟读扇区
  • WAF绕过：你的payload一提交就被拦？这10个冷门技巧让它畅通无阻
  • 失控AI Agent如何滥用urlquery远程浏览器执行，抽取俄罗斯政府数据
  • 美国智库发布《AGI时刻》报告，探讨通用人工智能未来发展
  • 学网络安全必知会：密码学
  • 惊心！歹毒！遭遇【VS编译时触发后门】保卫战
  • 婉莹事件的三个理解误区
  • 「AI Agent安全+提示注入+开源SDK」Superagent拆解：给大模型应用装安全护栏
  • 科普：旧手机怎么保护隐私
  • 用友某产品最新版前台RCE
  • Meta 开源 Muse 个人桌面AI Agent 硬件，适配乐鑫ESP32、树莓派，可处理邮件、在线购物、规划行程、传感器数采，智能家居控制
  • 新华三H3C网关-安全设备sslvpn_client.php存在远程命令执行漏洞
  • 【10-06 】Claude 最新防A/封号方案(不保真)
  • 苹果将收紧macOS全磁盘访问权限，防范AI Agent隐私风险
  • Google内部AI安全Agent发现并实证500+XSS漏洞
  • Fortinet FortiMail 0Day漏洞已遭在野利用
  • SDC2026议题预告 | Storm: AI审计时代下，Fuzzing还有用吗？
  • 基于CVD的云手机定制与风控分析
  • Application跳转Bootloader，如何处理“外狗”？
  • 如何监控自动驾驶系统中AI算法的安全性
  • 电子取证初探——六种常见图片隐写
  • 当一生的坚守被认真地讲述，崔老不禁红了眼眶
  • 《神探之痕迹》张译人物特辑
  • “找名字的人”电影《神探之痕迹》主题曲
  • 记一次提示词注入实战：从AI智能助手失控到企业数据泄露的渗透测试复盘
  • 之前有个领导非常恶心，专门招985、211高学历的人，然后故意给人家打差绩效，就为了表明自己比985的优秀，网友：这种恶人为什么会被重用
  • 从 union 读到路径，到写进一句话：SQL 注入 getshell 这条链，到底卡在哪
  • 2026年AI+网络安全专家班培训（web安全、内网渗透、代码审计、云安全、安全开发等）
  • 马哥教育AI+SRE云计算运维培训，0基础也能学，月入过万不是梦！
  • HackMyVm靶场-Hommie靶场渗透实战
  • Black Hat USA 2026：AI时代的数字私掠者
  • Android逆向工程终极利器：Claude Code + jadx双引擎，混淆代码也能秒变明文
  • 百度员工：国庆回家，发现老家公检法的地位真的高，A8的级别在这类人面前点头哈腰，自己只能在角落
  • ShinyHunters勒索组织被FBI拘留事件技术溯源分析
  • 我看"王总"就很迷茫——“迷茫啊，特别迷茫”（两个年轻人的一次聊天）
  • CNNIC 真的不可信吗？来看看那些重创互联网信任的 CA 安全事故
  • PIIFileScanner公民个人信息检测匹配系统
  • PIIFileScanner检测匹配工具与服务后端交互数据详解
  • 蜜网将成为抵御AI黑客的必备防线
  • 只要你上班用电脑，绝对用得到这13个Skills
  • AI取证实战第22期：小米手机微信聊天数据恢复
  • 网络安全每周简报｜本周50大网络安全热点新闻
  • 2026年福建省大学生数据安全大赛火热报名中！
  • 2026年第23届信息安全与对抗技术竞赛通知–智能安全赛
  • 关于举办首届江苏省“人工智能+”创新应用大赛的通知
  • 赛查查喊你来投票！ | 投出你最喜欢的安全赛事
  • 君说安全｜非网络安全专业能入行吗？我的结论是：有兴趣，大胆进
  • 基于混淆算法的端侧大模型保护机制研究
  • 《人工智能安全治理框架3.0》：完善人工智能安全治理，推动技术安全可控发展
  • 【一个人的SOC】(00) 中小企业为什么需要自己的安全运营中心
  • 从蔡康永事件到安全
  • 一个运维"宝藏项目"火了 但它真正戳中的，是全行业的痛
  • 红队工程师的盲区： 你手里的C2，可能正在反控你
  • 求职面试竟是陷阱！ContagiousInterview传染性面试，VSCode打开项目即中招，受害者还会无意识传播恶意仓库
  • MBA的最后一课
  • 当间谍软件CEO称：我们看不见客户在监视谁也没有终止开关
  • 屏幕背后的护国者：揭秘中国红客，网络世界里的无名卫士
  • 全生命周期需求视角的高质量数据集供给研究
  • 0xGame 2026 wp
  • 大厂面试官：你的 Agent 跑着跑着 token 爆了，怎么解决？ (网安面试题图解)
  • 千人大帮会｜网盘资源终身独享｜CISP/PTE考证底价｜18年网安大佬带队
  • 震惊！百度开源这款安全扫描工具，精准定位十大Web漏洞，误报率极低！
  • 银狐C2追踪连续三天静止：深度休眠还是监测盲区？
  • 我渗透到一个黑客集团：一年半时间，35xa0万美元换xa07500xa0万美元冻结
  • 攻击者将 ChatGPT 自定义 GPT 变成恶意软件传播诱饵
  • Dell System Update 5个漏洞：最高严重度9.6，可被用于root权限代码执行
  • Adminer 5.4.2 三个0-day漏洞分析与修复建议
  • 汽车安全入门04_攻击面建模：从物理接触到远程
  • WebSocket协议漏洞02_IDOR越权与DoS拒绝服务实战
  • 代码能跑就敢上线？这款大厂开源AI审计工具，帮你堵死90%隐形漏洞
  • 浅谈AI数字底座安全保障体系建设
  • 账户删除绕过漏洞
  • 硬件 key 授权机制的信任模型缺陷与加固思路
  • 全国职业院校技能大赛《信息安全管理与评估》真题+题目+解析
  • 公钥能换，分支能改：为什么本地签名验签授权必然被绕过
  • 你看不到的，我们守着
  • CVE-2026-61500 | Rejetto HFS会话伪造导致远程代码执行（POC）
  • 小样儿，以为把验证搬上服务端，就高枕无忧了？
  • 你对AI说的“悄悄话”，可能不是秘密
  • N/A | ShowDoc未授权PHP代码注入漏洞（POC）
  • 过节宅家打游戏：《波斯王子:失落王冠》
  • EasyCVR视频管理平台adduser未授权添加用户漏洞
  • (9.3分) CVE-2026-61500：HFS文件服务器认证绕过RCE
  • Java安全—EL 表达式注入
• Microsoft Security Blog
  • CISO perspectives on managing vulnerability risks in the age of AI
• Recent Commits to cve:main
  • Update Tue Oct 6 12:13:24 UTC 2026
• Kitploit — Hacking, PenTest, and Cybersecurity Tools for Your Security Arsenal!
  • Decretum
  • vedas-signatures
  • splitting-the-email-atom
  • h2spacex
  • Snitch
  • KDU
  • ipsw-diffs
  • HDI-GraphRAG-Attack
  • SafeGuard
  • ActiveScanPlusPlus
  • WS_RaceCondition_PoC
  • BinDiffHelper
  • gryph
  • worldmonitor — Updated!
  • caeruleus
  • Python-Shellcode-Anydesk-Apc-injection-Remote-IP-Address-Analysis
  • ABrake27
  • frida-gum
  • promptfoo v0.124.0
  • ultrasploiter
  • benign-instruction-bench
  • sliver-gui
  • trufflehog v3.98.0
  • rea
• Pentest Blog
  • hacked by trenggalek6etar
• Horizon3
  • CVE-2026-21589 | Atlassian Data Center Products Unauthenticated Arbitrary File Read Vulnerability
• Binary Ninja
  • Reversing Engineering a Captive Portal
• PortSwigger Research
  • The model isn't cooperating
• Reverse Engineering
  • Reverse engineering a Magelight XP glitch in Skyrim
  • I made this javascript deobfuscator for https://github.com/javascript-obfuscator/javascript-obfuscator
  • Inside the Windows I/O Manager : Deep dive into how read I/O requests are initialized.
  • Autel EVO Nano+: getting a root shell | 0xD34D’s Space
  • iOS 27.2 beta: silent hardening in ImageIO ASTC decoder — structural, zero-click candidate
  • Testing my Browser Based SaavyCAN-style Program
  • Is it possible to decompile the source code of Mario Kart Arcade GP DX with AI?
• Malwarebytes
  • ASOS “hackers” send push notifications to customers
  • Facebook Marketplace scam uses your name and number
  • Domino’s customers targeted in credential stuffing attacks
• rtl-sdr.com
  • ESPsoup: Turn an ESP32-C5 into a 2.4 & 5 GHz Pocket Scanner with a Connected Phone or PC
  • ChronAlert: Live Dashboard for Radio, Weather, Alerts with RTL-SDR Integration for APRS, ADS-B and AIS
• Whwlsfb's Tech Blog
  • hacked by trenggalek6etar
• 看雪学苑
  • SDC2026议题预告 | Storm: AI审计时代下，Fuzzing还有用吗？
  • 基于CVD的云手机定制与风控分析
• 安全小飞侠
  • 看我如何使用Trae全程用AI开发了Dota游戏
• 白帽Wiki - 一个简单的wiki
  • [2026]一个token问题导致tsecbench-CTF的A-18(CloudFunc)做不出来
• 黑鸟
  • 德国前联邦情报局局长涉嫌叛国性间谍活动被捕
• 安全分析与研究
  • 自主漏洞发现与利用生成
• 安全圈
  • 【安全圈】AI垃圾报告泛滥成灾：Google暂停开源软件漏洞赏金
  • 【安全圈】黑客骨干约旦落网：ShinyHunters核心遭捕，FBI深入撕开7000万黑产链
  • 【安全圈】Atlassian曝9.3分严重漏洞：未授权遍历直读8款企业核心系统文件
• 极客公园
  • 不写代码的人，正在涌入 GitHub
  • OpenAI 宣布「28 天计划」，持续改进 Codex、Work；TikTok 上线一系列 AI 电商功能；蓝色起源公布月球「动力塔计划」 | 极客早知道
• TrustedSec
  • Logging is a Discipline, Not a Switch
• bellingcat
  • Groups Associated With India’s Far-Right Receive Millions Through US Donor-Advised Funds
• LockBoxx
  • OpenAI’s Artist Rules Are Easier to Evade Than to Explain
• Schneier on Security
  • Possible Vulnerability in Apple’s Automatic Reboot
• NETRESEC Network Security Blog
  • Stop Feeding the SOC Garbage
• SANS Internet Storm Center, InfoCON: green
  • More RMM Tools In the Wild, (Tue, Oct 6th)
  • ISC Stormcast For Tuesday, October 6th, 2026 https://isc.sans.edu/podcastdetail/10124, (Tue, Oct 6th)
• ICT Security Magazine
  • Sicurezza digitale e vulnerabilità al phishing: profilo demografico e abitudini
• Over Security
  • Ninja Forms plugin flaw exploited to hack WordPress sites
  • Hackers exploit 32 zero-days on first day of Pwn2Own Ireland
  • Atlassian warns of critical file-access flaw in Jira, Confluence
  • ASOS confirms data breach after “HACKED” in-app notifications
  • Alleged ATM malware creator appears in Nebraska court after arrest
  • South Korean officials believe AI agents were used to hack several banks
  • Fake ChatGPT, Gemini Sites steal advertising accounts, MFA codes
  • Osaka Metropolitan University cancels classes after suspected ransomware attack
  • L’evoluzione del ransomware: come l’AI potenzia le negoziazioni
  • How to secure RMM software: 8 controls MSPs should test
  • Arabic Streaming Websites ClickFix Campaign
  • The Hidden Risks of Affiliate Brand Bidding
  • ClickFix campaign in Ukraine compromises over 100 websites to spread Lunex malware
  • Today I learned: Python's .start Files as a Persistence Mechanism
  • EXCLUSIVE: Accenture contractor removed from FBI following damaging data breach, sources say
  • Shares in British clothing company ASOS dive after hackers apparently send push notification
  • Wikimedia: Rogue OpenAI agents behind unauthorized Wikipedia edits
  • IronChain Ransomware Threatens Businesses with Permanent Data Loss and Costly Downtime
  • Cybertech Europe 2026: AI e resilienza ridisegnano la cyber security europea
  • Torna la “truffa della ballerina”: falsi concorsi su WhatsApp per sottrarre gli account
  • Dipendenza cognitiva dall’AI: quando deleghiamo alla macchina anche il dubbio
  • From Detonation to Detection: The Sandbox Now Writes the Rules
  • Nikkei discloses breaches of employees’ Microsoft, Google email accounts
  • Super Intelligence: The Rise of Super-Intelligent Threats
  • Servizi di ascolto e GDPR: dalla DPIA alla gestione del rischio, come costruire l’accountability
  • Engineer sentenced for locking over 3,000 devices on employer network
  • 8.8 Million People Affected in Major Denmark Data Breach
  • Wikimedia Finds Unauthorized OpenAI Agent Activity Across Its Platforms
• The Hacker News
  • Fake ChatGPT, Gemini, and Claude Ad Portals Capture Credentials and MFA Codes
  • Linux Backdoors Impersonate Email Security Tools to Evade Detection in Korea and Taiwan
  • LibreOffice and OpenOffice Flaws Let Malicious Spreadsheets Run Code Without Macro Warnings
  • Wikimedia Says OpenAI Agents Tried to Compromise Etherpad and Use Wiki Tools as Proxies
  • Welcome to the Jungle: What We Found Inside 15,465 Public MCP Servers
  • Google Pauses OSS Product Bug Bounty Rewards After Surge in Invalid Automated Reports
  • Critical Atlassian Flaw Lets Unauthenticated Attackers Read Known Files Across 8 Products
  • FBI Removes Accenture Contractor After Patch Failure Led to ShinyHunters Breach
  • Denmark Says Attackers Accessed CPR Data for 8.8 Million People via Company Account
  • ClickFix Smuggles Payloads Through Browser Cache to Bypass Windows Run Limits
• KitPloit - PenTest Tools!
  • Decretum
  • vedas-signatures
  • splitting-the-email-atom
  • h2spacex
  • Snitch
  • KDU
  • ipsw-diffs
  • HDI-GraphRAG-Attack
  • SafeGuard
  • ActiveScanPlusPlus
  • WS_RaceCondition_PoC
  • BinDiffHelper
  • gryph
  • worldmonitor — Updated!
  • caeruleus
  • Python-Shellcode-Anydesk-Apc-injection-Remote-IP-Address-Analysis
  • ABrake27
  • frida-gum
  • promptfoo v0.124.0
  • ultrasploiter
  • benign-instruction-bench
  • sliver-gui
  • trufflehog v3.98.0
  • rea
• Security Affairs
  • CVE-2026-96940: Microsoft Fixes Exchange Server Flaw For Which Exploitation Is More Likely
  • FBI Drops Accenture Contractor After Sensitive Data Breach
  • Dell Urges Customers to Patch Critical DSU Flaw That Can Give Attackers Root Access
  • ClingSTUN Linux Backdoor Abuses Public STUN Infrastructure
• Full Disclosure
  • [0day-rubbish] StreamSets Transformer 3.17.0 auth-mode none fallback and un-sandboxed ScalaDTransform execution to container root (8.1 primary)
  • [0day-rubbish] RCDevs WebADM 2.4.14 authenticated log viewer sid command injection to webadm uid 999 code execution (7.2)
  • [0day-rubbish] Maian Cart 3.8 addBanners unrestricted banner upload to PHP webshell and administrator command execution (7.2 primary, PR:H)
  • [0day-rubbish] IPConfigure Orchid VMS 26.3.0 authenticated DNF repository GPG-key property command injection to root (7.2)
  • [0day-rubbish] Circutor LineEds 24.11.14-r0 unauthenticated pwrstudio events.xml shellExecute command injection (9.8)
  • [0day-rubbish] Asustor ADM 3.5.9.RWM1 (AS602T) music.cgi act=live stored-filename command injection reaching system() (8.8 primary, PR:L)
  • [0day-rubbish] Advantech WebAccess Node 9.2.3 unauthenticated CrystalRpt.aspx file upload and path traversal to code execution in w3wp.exe (9.8)
  • SEC Consult Research 20261001 :: Arbitrary Email sender spoofing in Apple iCloud mail
  • SEC Consult SA-20260924-0 :: Multiple Vulnerabilities in Paessler PRTG Network Monitor #CVE-2026-4637 #CVE-2026-4638
• Security Weekly Podcast Network (Audio)
  • Typing, textGrain, NetScaler, Fortinet, Copilot, LibreOffice, Dots, Aaran Leyland - SWN #622
  • Getting Granular with Access, Attributes, and Intent - Alex Olivier - ASW #403
• SEI Blog
  • Navigating Program Complexity with the Waypoints Framework
• www.theregister.com - Articles
  • Anthropic reconfigures its cool kids security program
  • Trump Mobile customers' data dumped - and some never even received their gold device
  • Microsoft extends the Outlook naughty step with two more file types
  • Zombie instructions on carefully constructed web pages could trick GitHub Copilot CLI into sharing secrets
  • Asos app delivers a data leak threat instead of fast fashion
  • Denmark's ID register spills more people's details than the country has residents
  • Legacy sign-on service comes back to bite school software provider Bromcom
  • Atlassian warns of critical file access flaw in its datacenter products
  • Security researcher claims they found KVM guest-host escape flaw
• TorrentFreak
  • Denuvo Asks Court to Unmask Game Cracker ‘voices38’, Reveals Crypto Transactions
• Project Zero
  • How to fix a bug in a fix