MFSA 2026-49Security Vulnerabilities fixed in Firefox for iOS 151.0
MFSA 2026-49Security Vulnerabilities fixed in Firefox for iOS 151.0
2026-07-12 归档
今日共收录 877 条安全情报，包含 71 个 CVE，468 篇安全通告，以及 78 篇研究论文。
← 返回归档列表MFSA 2026-52Security Vulnerabilities fixed in Firefox for iOS 151.1
MFSA 2026-52Security Vulnerabilities fixed in Firefox for iOS 151.1
MFSA 2026-53Security Vulnerabilities fixed in Firefox for iOS 151.2
MFSA 2026-53Security Vulnerabilities fixed in Firefox for iOS 151.2
MFSA 2026-55Security Vulnerabilities fixed in Focus for iOS / Klar 151.3.1
MFSA 2026-55Security Vulnerabilities fixed in Focus for iOS / Klar 151.3.1
MFSA 2026-56Security Vulnerabilities fixed in Firefox for iOS 152.0
MFSA 2026-56Security Vulnerabilities fixed in Firefox for iOS 152.0
iOS 26.5.2 and iPadOS 26.5.2
iOS 26.5.2 and iPadOS 26.5.2
MFSA 2026-65Security Vulnerabilities fixed in Firefox for iOS 152.3
MFSA 2026-65Security Vulnerabilities fixed in Firefox for iOS 152.3
【大模型服务平台 TokenHub】&【智能体开发平台】& 【云开发 CloudBase】 关于腾讯云 MiniMax-M2.5 模型下线及切换升级的通知
【大模型服务平台 TokenHub】&【智能体开发平台】& 【云开发 CloudBase】 关于腾讯云 MiniMax-M2.5 模型下线及切换升级的通知
Stable Channel Update for ChromeOS / ChromeOS Flex
The ChromeOS Stable channel is being updated to OS version 16667.62.0 (Browser version 149.0.7827.238) for most ChromeOS
TRACE: A Two-Channel Robust Attribution Watermark via Complementary Embeddings for LLM-Agent Trajectories
大型语言模型（LLM）智能体通常通过转售商（reseller）分发给最终用户。转售商可能对开发者的智能体进行品牌重塑，或替换为更便宜的模型。当溯源发生争议时，归责依据的是轨迹日志（trajectory log），即工具调用、观察结果和执行动
Multi-Agent Firewall Architecture for Privacy Protection of Sensitive Data in Interactions with Language Models
本文针对大语言模型（LLM）集成到工作流中带来的敏感数据泄露风险，提出了一种开源的、面向用户的多智能体防火墙架构。该架构由浏览器扩展和代理组成，能够全面拦截HTTP(S)和WebSocket通信流量。核心是一个灵活的多智能体管道，采用确定性
Token-Flow Firewall: Semantic Runtime Auditing for Persistent AI Agents
本文针对持久化AI agent（如具备长期记忆、可调用工具的多轮交互系统）面临的安全挑战，提出了一套运行时语义防火墙框架TokenWall。传统单轮聊天助手的防御方法无法应对持久化agent中通过自然语言token流（包括记忆更新、工具参数
From Legacy Documentation to OSCAL: An MCP-Based Agent Pipeline for Threat-Informed Continuous Compliance in Critical Infrastructure
该论文针对关键基础设施中运营技术（OT）环境难以主动扫描但风险评估和合规性仍需实时反馈的问题，提出了一种基于模型上下文协议（MCP）的多智能体管道，用于将自然语言系统描述转化为经过源验证的知识图谱和符合NIST OSCAL格式的审计就绪工件
Multimodal Unlearning Across Vision, Language, Video, and Audio: Survey of Methods, Datasets, and Benchmarks
随着视觉语言模型（VLM）、扩散模型（DM）、大语言模型（LLM）和音频基础模型（AFM）的广泛应用，这些多模态基础模型可能会从训练数据中无意编码敏感、受版权保护、有偏见或不安全的跨模态关联。在收到删除请求或政策更新后重新训练通常不切实际，
PraisonAI before 1.6.78 contains a remote code execution vulnerability in CodeAgent._execute_python() that executes LLM-generated Python code without AST validation, import rest...
CVE-2026-61447
CVE-2026-61447 是 PraisonAI 1.6.78 之前版本中存在的一个严重远程代码执行漏洞。该漏洞位于 CodeAgent._execute_python() 方法中，该方法直接执行由大语言模型（LLM）生成的 Pytho
PraisonAI before 4.6.78 contains arbitrary file write and command execution vulnerabilities in the AICoder component due to missing path validation and command sanitization in L...
CVE-2026-61445
PraisonAI 在 4.6.78 之前的版本中存在严重安全漏洞，位于 AICoder 组件中。该漏洞源于 LLM 工具调用时缺乏路径验证和命令清理，允许攻击者通过聊天界面注入恶意提示。攻击者可以利用此漏洞将任意文件写入文件系统的任意位置
Cloud & Application Security
Cloud & Application Security
From Scanner to Stealer: Inside the trivy-action Supply Chain Compromise
From Scanner to Stealer: Inside the trivy-action Supply Chain Compromise
Next-Gen Identity Security
Next-Gen Identity Security
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