2026-09-05 归档

今日共收录 1726 条安全情报,包含 665 个 CVE,657 篇安全通告,以及 120 篇研究论文。

← 返回归档列表
安全公告2026-09-05 发布INFO

Improper neutralization of CRLF sequences in IXON VPN Client before version 1.4.7 allows an...

CVE-2026-75925

Improper neutralization of CRLF sequences in IXON VPN Client before version 1.4.7 allows an attacker to execute commands

历史摘要 · 未重新核验 · GitHub Advisory

安全公告2026-09-04 发布INFO

OpenChoreo: cluster-gateway internal proxy performs no caller authentication and is not read-only — data-plane Secret disclosure and arbitrary Kubernetes mutation

CVE-2026-73842

### Summary The OpenChoreo control-plane cluster-gateway exposes internal management APIs (`/api/proxy/`, `/api/exec/`,

历史摘要 · 未重新核验 · GitHub Advisory

安全公告2026-09-03 发布INFO

Chromium: CVE-2026-84349 Use after free in Browser

CVE-2026-84349

CVE-2026-84349 是 Chromium 中浏览器组件(Browser)存在的一处释放后使用(Use after free)漏洞。Microsoft Edge(基于 Chromium)会集成 Chromium 的上游代码,因此该漏

历史摘要 · 未重新核验 · Microsoft MSRC

学术论文2026-09-05 发布INFO

GPU Travelling: Efficient Confidential Collaborative Training with TEE-Enabled GPUs.

本文提出了一种名为 GPU Travelling 的新型机制,旨在解决机密协作机器学习(Confidential Collaborative ML)中因模型和数据集规模快速增长而导致的通信开销过大的问题。在传统的多方联合训练场景中,多个互不

历史摘要 · 未重新核验 · ACM CCS

学术论文2026-09-05 发布INFO

Sok: Evaluating Jailbreak Guardrails for Large Language Models.

本文是一篇针对大语言模型(LLM)越狱防护机制的系统化知识(SoK)综述。研究背景在于:尽管LLM取得了显著进展,但其部署暴露了严重的安全漏洞,尤其是越狱攻击能够绕过模型的安全对齐机制。为应对这一威胁,防护栏(guardrails)——即用

历史摘要 · 未重新核验 · IEEE S&P

安全公告2026-09-05 发布INFO

Ivanti Endpoint Manager Mobile代码注入漏洞

根据知道创宇Seebug漏洞库收录的信息,存在一则关于Ivanti Endpoint Manager Mobile的代码注入漏洞公告。公告标题表明该产品存在代码注入风险,但当前公告中未提供具体受影响版本、CVE编号、技术细节或严重性分级。I

历史摘要 · 未重新核验 · 知道创宇 / Seebug

安全公告2026-09-05 发布INFO

In Ignition 8.1.53 and earlier, the Gateway "Create Project Role(s)" setting shipped blank, which...

CVE-2026-77393

In Ignition 8.1.53 and earlier, the Gateway "Create Project Role(s)" setting shipped blank, which permitted any authenti

历史摘要 · 未重新核验 · GitHub Advisory

安全公告2026-09-04 发布INFO

SiYuan: Non-administrator responses from /api/system/getConf omit three secrets that the configuration-export path explicitly strips, disclosing the session-cookie signing key a...

CVE-2026-72793

**CVE:** This vulnerability corresponds to [CVE-2026-72793](https://nvd.nist.gov/vuln/detail/CVE-2026-72793). ### Summa

历史摘要 · 未重新核验 · GitHub Advisory

安全公告2026-09-04 发布INFO

SiYuan: The session-cookie signing key (Conf.CookieKey) is returned to anonymous readers by /api/system/getConf

CVE-2026-72794

**CVE:** This vulnerability corresponds to [CVE-2026-72794](https://nvd.nist.gov/vuln/detail/CVE-2026-72794). ### Summa

历史摘要 · 未重新核验 · GitHub Advisory

安全公告2026-09-04 发布INFO

SurrealDB: Custom API route lets authenticated callers override namespace/database scope via URL path

CVE-2026-63735

An authenticated user scoped to one namespace/database could invoke a custom API (`DEFINE API`) belonging to a different

历史摘要 · 未重新核验 · GitHub Advisory

威胁情报2026-09-04 发布INFO

Critical Citrix NetScaler auth bypass now leveraged in attacks

CVE-2026-19490

2026年9月4日,据BleepingComputer报道,安全情报公司Previdian指出攻击者已在野外针对Citrix NetScaler中一个严重级别的身份验证绕过漏洞(CVE-2026-19490)展开利用。该漏洞允许攻击者绕过身

历史摘要 · 未重新核验 · BleepingComputer

威胁情报2026-09-04 发布INFO

How to secure edge AI in customer-owned environments

本文是微软安全团队于2026年9月4日发布的一篇技术指南,主题为在客户自有的边缘计算环境中安全部署和管理AI系统。随着AI推理逐渐从云端下沉到边缘设备,组织往往需要在自身控制的基础设施上运行AI模型,并处理敏感数据、凭据和模型资产。微软指出

历史摘要 · 未重新核验 · Microsoft Threat Intelligence

安全公告2026-09-04 发布INFO

CodeWhale: Argument Injection in `git_show` Tool Allows Arbitrary File Write Without Approval

CVE-2026-75913

### Maintainer resolution The CodeWhale maintainers validated this report. The affected package ranges are recorded in

历史摘要 · 未重新核验 · GitHub Advisory

学术论文2026-09-05 发布INFO

Web Application Vulnerability Repair Via Context-Aware Fault Localization and Directed Differential Fuzzing.

本文针对Web应用程序漏洞修复过程中存在的关键挑战:漏洞定位困难、修复补丁可能引入新问题或修复不完整等。作者提出了一种结合上下文感知的故障定位(Context-Aware Fault Localization)与定向差异模糊测试(Direc

历史摘要 · 未重新核验 · IEEE S&P

学术论文2026-09-05 发布INFO

SWIFT: Super-fast and Robust Privacy-Preserving Machine Learning.

本文提出 SWIFT,一个面向安全外包计算(SOC)范式的鲁棒隐私保护机器学习(PPML)框架。在 SOC 中,计算被外包给一组功能强大的服务器,按使用付费提供服务。现有 PPML 框架通常仅提供公平性(fairness),即恶意方可以阻止

历史摘要 · 未重新核验 · USENIX Security

安全公告2026-09-05 发布INFO

Citrix NetScaler 内存泄漏(CVE-2025-5777)

CVE-2025-5777

Citrix NetScaler 内存泄漏(CVE-2025-5777)

历史摘要 · 未重新核验 · 知道创宇 / Seebug

安全公告2026-09-03 发布INFO

Chromium: CVE-2026-84359 Information leak in Skia

CVE-2026-84359

This CVE was assigned by Chrome. Microsoft Edge (Chromium-based) ingests Chromium, which addresses this vulnerability.

历史摘要 · 未重新核验 · Microsoft MSRC

安全公告2026-09-03 发布INFO

Chromium: CVE-2026-84358 Improper privilege management in Downloads

CVE-2026-84358

This CVE was assigned by Chrome. Microsoft Edge (Chromium-based) ingests Chromium, which addresses this vulnerability.

历史摘要 · 未重新核验 · Microsoft MSRC

安全公告2026-09-03 发布INFO

Chromium: CVE-2026-84357 Improper input validation in Omnibox

CVE-2026-84357

This CVE was assigned by Chrome. Microsoft Edge (Chromium-based) ingests Chromium, which addresses this vulnerability.

历史摘要 · 未重新核验 · Microsoft MSRC

安全公告2026-09-03 发布INFO

Chromium: CVE-2026-84356 UI misrepresentation in FullScreen

CVE-2026-84356

This CVE was assigned by Chrome. Microsoft Edge (Chromium-based) ingests Chromium, which addresses this vulnerability.

历史摘要 · 未重新核验 · Microsoft MSRC