2026-07-21 归档

今日共收录 1301 条安全情报,包含 465 个 CVE,462 篇安全通告,以及 91 篇研究论文。

← 返回归档列表
安全公告2026-07-21 发布INFO

Use after free in GPU in Google Chrome on Android prior to 150.0.7871.128 allowed a remote...

CVE-2026-15900

Use after free in GPU in Google Chrome on Android prior to 150.0.7871.128 allowed a remote attacker to potentially perfo

历史摘要 · 未重新核验 · GitHub Advisory

安全公告2026-07-21 发布INFO

AVideo before 29.0 contains an incomplete fix for CVE-2026-45578 where execAsync() re-wraps...

CVE-2026-64625

AVideo before 29.0 contains an incomplete fix for CVE-2026-45578 where execAsync() re-wraps escaped commands in double-q

历史摘要 · 未重新核验 · GitHub Advisory

安全公告2026-07-20 发布INFO

@astrojs/node: Backslash-prefixed paths not recognized as internal by trailing-slash redirect

CVE-2026-59730

### Impact With `trailingSlash: 'always'` configured, the `@astrojs/node` standalone server's static file handler appen

历史摘要 · 未重新核验 · GitHub Advisory

安全公告2026-07-20 发布INFO

Astro: Cross-site scripting via unescaped transition:* directive values on hydrated islands

CVE-2026-59727

### Summary When a `transition:persist`, `transition:scope`, or `transition:persist-props` directive is applied to a cl

历史摘要 · 未重新核验 · GitHub Advisory

安全公告2026-07-20 发布INFO

File Browser: Colliding username normalization gives two users the same home directory

CVE-2026-62685

## Summary FileBrowser confines each user to a *scope*: a home directory that acts as the boundary for everything they

历史摘要 · 未重新核验 · GitHub Advisory

安全公告2026-07-20 发布INFO

File Browser: Share API exposes the password hash and bypass token

CVE-2026-62684

## Summary When a user creates a password-protected share or lists existing shares, the JSON response includes the full

历史摘要 · 未重新核验 · GitHub Advisory

安全公告2026-07-20 发布INFO

File Browser: Archive builder turns backslash filenames into path traversal (zip-slip)

CVE-2026-62843

## Summary The fix for `GHSA-gxjx-7m74-hcq8` / `CVE-2026-54093` (shipped in v2.63.6) added a `strings.ReplaceAll(nameIn

历史摘要 · 未重新核验 · GitHub Advisory

安全公告2026-07-20 发布INFO

node-tar: Process crash via PAX numeric path type confusion

CVE-2026-59871

### Summary A crafted 2.5KB tar archive crashes any Node.js process that extracts it. The PAX header parser coerces all

历史摘要 · 未重新核验 · GitHub Advisory

安全公告2026-07-20 发布INFO

Socket.IO: Engine.IO Polling Transport Connection Exhaustion

CVE-2026-59725

### Impact An unauthenticated remote attacker can cause a denial of service in affected versions of **engine.io** by op

历史摘要 · 未重新核验 · GitHub Advisory

安全公告2026-07-20 发布INFO

Mistune toc / TableOfContents directive: heading IDs use predictable `toc_N` numbering with no slugification, allowing collision with attacker-controlled `id="toc_N"` content

CVE-2026-59930

## Summary **Type:** Predictable identifier generation. The `toc` plugin and `TableOfContents` directive both default t

历史摘要 · 未重新核验 · GitHub Advisory

安全公告2026-07-20 发布INFO

Mistune: XSS via percent-encoded javascript URI bypass in safe_url()

CVE-2026-59923

### Summary An XSS vulnerability in Mistune allows bypassing of safe_url() protections via percent-encoded javascript UR

历史摘要 · 未重新核验 · GitHub Advisory

安全公告2026-07-20 发布INFO

VSee Clinic 7.1.26 and VSee Clinic API 1.3.0 exposes cleartext SFTP credentials in the HTTP...

CVE-2026-13380

VSee Clinic 7.1.26 and VSee Clinic API 1.3.0 exposes cleartext SFTP credentials in the HTTP responses of three unauthent

历史摘要 · 未重新核验 · GitHub Advisory

学术论文2026-07-20 发布INFO

GARAGE: Characterizing the Automation Boundary in LLM-based Attack Graph Generation

GARAGE 是一个基于检索增强生成(RAG)的框架,旨在弥合当前车辆安全领域自动工具在处理非结构化网络威胁情报(CTI)和汽车特定架构细节方面的不足。该框架将碎片化的 CTI 转化为可操作的、领域特定的知识库,用于自动生成攻击图。具体地,

历史摘要 · 未重新核验 · arXiv cs.CR

学术论文2026-07-19 发布INFO

Measuring and Evaluating the Performance of Generative AI Models for Scam Detection

本文研究了大型语言模型(LLMs)在在线诈骗检测中的适用性。在线诈骗造成严重的经济和个人损失,目前基于LLM的检测系统已被集成到邮件网关、浏览器扩展和欺诈监控面板等安全产品中。然而,人们普遍认为这些模型普遍适用于诈骗检测,但缺乏系统性验证。

历史摘要 · 未重新核验 · arXiv cs.CR

学术论文2026-07-17 发布INFO

One Modality to Forget Them All: Enhancing Cross-Modal Unlearning in Vision-Language Models

该论文首次系统研究了视觉-语言模型(VLM)中遗忘(unlearning)的跨模态迁移问题。现有机器学习遗忘技术主要用于移除大型语言模型中的危险知识,但VLM同时处理文本和视觉输入,遗忘在一个模态中是否会自动迁移到另一个模态是一个关键安全问

历史摘要 · 未重新核验 · arXiv cs.CR

威胁情报2026-07-21 发布INFO

Cloud & Application Security

Cloud & Application Security

历史摘要 · 未重新核验 · CrowdStrike

威胁情报2026-07-21 发布INFO

From Scanner to Stealer: Inside the trivy-action Supply Chain Compromise

From Scanner to Stealer: Inside the trivy-action Supply Chain Compromise

历史摘要 · 未重新核验 · CrowdStrike

威胁情报2026-07-21 发布INFO

Next-Gen Identity Security

Next-Gen Identity Security

历史摘要 · 未重新核验 · CrowdStrike

威胁情报2026-07-21 发布INFO

CrowdStrike Named an Innovation and Growth Leader in the 2026 Frost Radar™: Cloud and Application Runtime Security

CrowdStrike Named an Innovation and Growth Leader in the 2026 Frost Radar™: Cloud and Application Runtime Security

历史摘要 · 未重新核验 · CrowdStrike

威胁情报2026-07-21 发布INFO

CrowdStrike Expands Identity Leadership with OpenID and IDPro

CrowdStrike Expands Identity Leadership with OpenID and IDPro

历史摘要 · 未重新核验 · CrowdStrike