Azure Arc Elevation of Privilege Vulnerability
CVE-2026-69399
Azure Arc Elevation of Privilege Vulnerability
2026-09-18 归档
今日共收录 2436 条安全情报，包含 1258 个 CVE，779 篇安全通告，以及 100 篇研究论文。
← 返回归档列表Improper limitation of a pathname to a restricted directory ('path traversal') in Azure Arc...
CVE-2026-70009
Improper limitation of a pathname to a restricted directory ('path traversal') in Azure Arc allows an unauthorized attac
Authentication bypass by spoofing in Microsoft Dataverse allows an unauthorized attacker to...
CVE-2026-77903
Authentication bypass by spoofing in Microsoft Dataverse allows an unauthorized attacker to elevate privileges over a ne
Improper access control in Azure Logic Apps allows an unauthorized attacker to elevate privileges...
CVE-2026-83944
Improper access control in Azure Logic Apps allows an unauthorized attacker to elevate privileges over a network.
Missing authentication for critical function in Azure AI Foundry allows an unauthorized attacker...
CVE-2026-85889
Missing authentication for critical function in Azure AI Foundry allows an unauthorized attacker to elevate privileges o
Server-side request forgery in Omnibox in Google Chrome on on Android prior to 153.0.8010.52...
CVE-2026-93384
Server-side request forgery in Omnibox in Google Chrome on on Android prior to 153.0.8010.52 allowed a remote attacker l
Use after free in Dawn in Google Chrome on on Android prior to 153.0.8010.52 allowed a remote...
CVE-2026-93374
Use after free in Dawn in Google Chrome on on Android prior to 153.0.8010.52 allowed a remote attacker to potentially ex
Buffer overflow in WebGL in Google Chrome on on Android prior to 153.0.8010.52 allowed a remote...
CVE-2026-93372
Buffer overflow in WebGL in Google Chrome on on Android prior to 153.0.8010.52 allowed a remote attacker to execute arbi
Cute Editor for ASP.NET 6.4 is vulnerable to reflected cross-site scripting caused by improper...
CVE-2021-3030
Cute Editor for ASP.NET 6.4 is vulnerable to reflected cross-site scripting caused by improper validation of the Theme G
Grav: Missing admin.super guard on core group blueprint access field allows admin.users operator to escalate to super-admin
CVE-2026-75837
## Summary The core Flex group blueprint `system/blueprints/user/group.yaml` (access field, lines 48-55) omits the `secu
Grav: Single invalid UTF-8 byte disables every rule in Security::detectXss(), bypassing the page-content XSS safety gate
CVE-2026-75834
## Vulnerability Details **Component**: getgrav/grav core **File**: `system/src/Grav/Common/Security.php` **Function**:
Grav: detectXss() misses an event-handler attribute after an unpaired quote in an unquoted attribute value, giving stored XSS
CVE-2026-75828
## Affected versions and vulnerable location - Confirmed on grav core at `78ebfc1` (tag 2.0.13). - Detector: `system/sr
@platejs/core HTML deserialization can trigger browser behavior during parsing
CVE-2026-88976
### Summary HTML strings passed to Plate's core deserialization APIs were parsed in the active document. Certain HTML a
react/http: A malformed HTTP chunked body can lead to a denial-of-service and peg the CPU
CVE-2026-84997
### Summary A malformed HTTP message using `Transfer-Encoding: chunked` can drive `React\Http\Io\ChunkedDecoder` into a
OpenTelemetry-Go: Exporter config logging may leak endpoint URLs in info logs
CVE-2026-81870
### Summary OpenTelemetry Go versions 1.5.0 through 1.44.0 can include trace exporter endpoint configuration in an inte
Grav: media_directory() Twig function allows filesystem path traversal and file content disclosure from sandboxed page content
CVE-2026-72697
## Summary The `media_directory()` Twig function is allow-listed for use in sandboxed, editor-authored page content (`s
MAGS: Multi-agent Auto-formalization Guarantees Safety for Agentic Outputs
该论文关注 LLM 编码智能体（coding agent）大规模生成复杂程序后，人工审查难以覆盖、进而引发安全与功能失效风险的问题。作者指出，常用的保障手段各有局限：模糊测试只能覆盖被触发的输入路径，静态分析受限于规则与误报，LLM-as-
InceptionRAG: Stealthy Poisoning Attack Against Retrieval-Augmented Generation
该论文研究检索增强生成（RAG）系统中的语料库投毒问题。RAG 通过外部知识库为大语言模型补充事实依据，但已有研究表明其语料可被污染。现有的投毒攻击多采用“单点显式注入”范式，即把完整的恶意载荷封装在单篇文档中，因此近期的防御机制（如对单文
Autonomy in Check: Governor-Mediated Adaptive Security at the Edge
论文关注网络边缘自适应安全中自动化规划器（规则控制器、学习策略、LLM辅助代理）将观测转为执行动作的风险。作者指出，一旦规划器能影响实时策略状态，仅靠语法有效性不足：基于不完整或被操纵观测产生的语义错误动作，可能被无法理解任务上下文的执行基
Towards TEE-Certified DP: Verifiable Differentially Private Training on Legacy GPUs
该论文研究的是一个长期被忽视但监管意义重大的问题：如何在训练完成后向外部验证方证明「模型确实按照所声称的差分隐私（DP）机制完成了训练」，而验证方无需接触任何私有训练数据。随着机器学习在敏感数据上的广泛部署，监管与合规对「可信 DP 训练」