Aggregator

A vulnerability described as critical has been identified in Adobe Experience Manager. The affected element is an unknown function of the component XML External Entity Parser. The manipulation results in xml external entity reference. This vulnerability was named CVE-2026-48359. The attack may be performed from remote. There is no available exploit.

A vulnerability, which was classified as problematic, was found in pyasn1 up to 0.6.3. The impacted element is the function decode of the component BER/CER/DER Decoder. Such manipulation of the argument RELATIVE-OID leads to resource consumption. This vulnerability is documented as CVE-2026-59885. The attack can be executed remotely. There is not any exploit available.

A vulnerability was found in TecharoHQ Anubis up to 1.26.0-pre0. It has been rated as problematic. This issue affects the function PathChecker.Check of the file lib/policy/checker.go of the component Policy. Performing a manipulation results in improper input validation. This vulnerability is cataloged as CVE-2026-62314. It is possible to initiate the attack remotely. There is no exploit available.

A vulnerability was found in xxl-job-admin 3.0.0. It has been declared as problematic. This vulnerability affects unknown code of the component HTTP Handler. Such manipulation leads to cross site scripting. This vulnerability is listed as CVE-2026-26719. The attack may be performed from remote. There is no available exploit.

A vulnerability was found in Dulwich up to 1.1.0. It has been classified as critical. This affects an unknown part of the file contrib/paramiko_vendor.py. This manipulation causes Remote Code Execution. This vulnerability is tracked as CVE-2026-38974. The attack is possible to be carried out remotely. No exploit exists.

A vulnerability was found in xxl-job 3.0.0 and classified as problematic. Affected by this issue is some unknown functionality of the component Endpoint. The manipulation results in cross-site request forgery. This vulnerability is identified as CVE-2026-26718. The attack can be executed remotely. There is not any exploit available.

本文记录 全程AI Unity IL2CPP 逆行改包全流程：静态导出 dump 定位 AutoRevival 死亡链，Gadget 脱机注入实现碰撞续跑、去闪屏、无限复活，附 AI 指令与实机验收，技术细节保留、样本脱敏。

写了一个扫描器，扫出 192 个告警，其中 189 个是假的。花了三周重写 diff 引擎，误报率从 95% 只降到 90%。后来发现问题的根源不是算法精度，而是"单次观测无法区分噪声与真实触发"。本文提出 MSCT（Multi-Sample Consistency Test）——用双 payload 加无害对照请求代替传统单次 diff，让扫描器在没有人工干预的情况下把误报率降到可接受范围。同时

当所有反向代理都在配置层防御攻击时，Varnish 把整个配置语言编译成了机器码。 这意味着攻击者多了一条从未被正视的入侵路径：编译器。 我们在 Varnish 6.6.1 上挖到了 CVE-2022-23959 的 PoC——但补丁没修对。官方补了 body 失败路径，攻击者走的却是 pipeline 残留路径。补丁覆盖的代码分支和实际攻击路径正交。 这是中文社区首次从 VCL 编译链视角

在 ELF 加载早期，.note.gnu.property 这样的“非代码段”元数据已经足以影响 IFUNC resolver 的执行路径；一次不改 .text 的 note 篡改，可能让同一个符号在运行时落到完全不同的实现上。

任务失效化蜜罐的价值不是追求一次性欺骗，而是让 Agent 的任务闭环在受控环境中持续偏航、消耗并最终失去有效方向。

本文揭示 JDK 21 虚拟线程（JEP 444）引入的一类隐蔽会话越权漏洞。JDK 21 的虚拟线程默认名称为空字符串，当业务代码在未显式命名的虚拟线程中读写以线程名为键的隔离表（多租户上下文、APM 链路追踪、自定义 MDC）时，所有虚拟线程会退化为共享同一条目，导致并发请求间发生身份串扰。

A vulnerability has been found in Wekan up to 9.45 and classified as critical. Affected by this vulnerability is the function getRequestIp of the file server/lib/headerLoginAuth.js of the component Header-Login. The manipulation of the argument HEADER_LOGIN_ID leads to improper authentication. This vulnerability is referenced as CVE-2026-55652. Remote exploitation of the attack is possible. No exploit is available.

本文验证了 JNA 在可写目录释放临时 native 库时的路径替换风险，通过轮询目录项并利用 ELF constructor 在动态链接阶段执行恶意代码。文章演示了利用 os.replace 替换 jna*.tmp 文件实现攻击的方法，并提出了固定私有释放目录及关闭运行时解包的防御方案。

A vulnerability, which was classified as very critical, was found in MaaAssistantArknights. Affected is an unknown function of the file .github/workflows/release-preparation.yml of the component Pull Request Title Handler. Executing a manipulation of the argument pull_request.title can lead to os command injection. The identification of this vulnerability is CVE-2026-55576. The attack may be launched remotely. There is no exploit available.

记一次 Rust 免杀 Loader 的折腾过程0x00 起因之前一直在用 Go 写 loader，千机那个项目搞了好几版，效果越来越差。最离谱的是有一次改了几个字符串重新编译，360 直接秒杀，连行为检测都没走到就被 QVM 静态干了。翻了一下 virustotal 上面 Go 编译的样本，发现只要是 Go 写的，不管功能多简单，360 基本都标红。原因也简单——Go 的 runtime 太大了

BroncoCTF所有pwn题的wp

A vulnerability, which was classified as critical, has been found in Wekan up to 9.36. This impacts an unknown function of the file server/permissions/cards.js of the component Permissions. Performing a manipulation of the argument boardId results in improper authorization. This vulnerability was named CVE-2026-55234. The attack may be initiated remotely. There is no available exploit.

还在手动敲 ADB 命令脱壳、在 JADX 里反复搜索硬编码密钥？本文带你用 MCP 协议将 ADB、一键脱壳技能、Frida 和 JADX 串联成自动化流水线。只需一句话，AI 就能完成从设备连接、脱壳到代码审计的全流程，自动挖掘硬编码凭证、不安全 API 和导出组件风险，最终输出一份专业的安全分析报告。无论你是安全研究员还是逆向新手，都能大幅提升效率，告别重复劳动。

免杀：Patch白程序