CVE-2026-100540 OpenClaw Feishu before 2026.8.1 Authentication Bypass via Disabled Account

OpenClaw Feishu before 2026.8.1 fails to validate whether a configured default account is disabled before selecting it for model tool operations. Attackers can exploit multi-account setups where a disabled default account retains credentials to read or modify Feishu resources through a revoked...

CVE-2026-100540 OpenClaw Feishu before 2026.8.1 Authentication Bypass via Disabled Account

OpenClaw Feishu before 2026.8.1 fails to validate whether a configured default account is disabled before selecting it for model tool operations. Attackers can exploit multi-account setups where a disabled default account retains credentials to read or modify Feishu resources through a revoked...

CVE-2026-100528 OpenClaw before 2026.8.1 Credential Disclosure via Provider Endpoint

OpenClaw npm package 'openclaw' before 2026.8.1 could send third-party provider credentials to the wrong endpoint. In affected versions, when a third-party provider uses an OpenAI-compatible API and the resolved model metadata lacks a concrete base URL, a pinned session that continues after a mod...

CVE-2026-100528

The OpenClaw npm package (openclaw ) versions prior to 2026.8.1 are vulnerable to a credential disclosure flaw. When using a third-party provider with an OpenAI-compatible API that lacks a concrete base URL in its metadata, a pinned session following a model configuration hot reload may retain th...

CVE-2026-77991

abraxaslabs.tech · github.com/abraxas · @abraxasnull · CVE-2026-77991 CVE-2026-77991 JEM - Joomla Event Manager 5.0.0 — joomlaeventmanager.net Joomla 扩展 - joomlaeventmanager.net - Joomla Event Manager ，以便会话文件名匹配。 备注： CVE-2026-77991：JemModelSource::save 将 jformsource File::write 到...

CVE-2026-100653: Use of Less Trusted Source

vLLM is an inference and serving engine for large language models. In versions from 0.22.1 through 0.28.0, the operator-supplied model revision pin --revision / --code-revision is not propagated to several Hugging Face artifact loads for the FunAudioChat and Tarsier2 architectures: the...

CVE-2026-100540: Incorrect Authorization

OpenClaw Feishu before 2026.8.1 fails to validate whether a configured default account is disabled before selecting it for model tool operations. Attackers can exploit multi-account setups where a disabled default account retains credentials to read or modify Feishu resources through a revoked...

Debian dsa-6517 : libnode-dev - security update

The remote Debian 13 host has packages installed that are affected by multiple vulnerabilities as referenced in the dsa-6517 advisory. - ------------------------------------------------------------------------- Debian Security Advisory DSA-6517-1 [email protected] https://www.debian.org/securit...

CVE-2026-100556: Incorrect Authorization

OpenClaw npm package openclaw versions = 2026.5.2 and 2026.8.1 contain an incorrect authorization vulnerability in WhatsApp group handling. A group sender who is admitted for ordinary messages but denied by commands.allowFrom or owner command authorization can issue the /new command to reset the...

gha-lab-61c59f4acb

Automated research artifact — not the upstream project. This repository is a disposable lab built by an automated harness for a master's thesis at Université Laval on reproducing published GitHub Actions workflow vulnerabilities. It is a verbatim snapshot of labring/FastGPT at commit...

Rasa-cve-exp

" Rasa-cve-exp" CVE-2021-42556 server-1.py 直接运行即可 python server-1.py 任意文件上传 PUT /model "modelserver":"url":"http://192.168.244.128:5000/stage1","waittimebetweenpulls":0 效果如下： CVE-2021-41127 server-1.py 默认模型名evilimportmodel.tar.gz 恶意模型demo version: "3.1" intents: - greet responses: uttergreet: -...

CVE-2026-5843

CVE-2026-5843 PoC 一个极简的 OCI 注册表，用于提供恶意模型，以利用 mlx-lm 中的 modelfile importlib 执行路径。当 Docker Model Runner 加载模型进行推理时，model.py 将以当前登录用户的身份在宿主机上执行。 要求： Docker Desktop ≤ 4.70.x（Apple Silicon）并启用 Model Runner 使用方法 1. Start the registry on the host python3 poccve20265843.py 2. From any container on the...

From Debugging to Code Execution: RCE in Microsoft DevLabs’ DebugMCP?

Introduction AI-assisted development has crossed a threshold. The question is no longer whether your IDE communicates with an AI, it's how deeply that AI is now wired into your local environment. The Model Context Protocol MCP is one of the plumbing behind this shift: a JSON-RPC-based protocol th...

CVE-2026-79721

A flaw was found in MLflow. This vulnerability allows a remote attacker to achieve arbitrary code execution on an end user's system. This occurs when a maliciously crafted model artifact is loaded by the MLflow platform, enabling the execution of unauthorized code. Mitigation Avoid loading,...

zombieguard

ZombieGuard Un escáner defensivo para la evasión de metadatos estructurales en archivos ZIP. ZombieGuard detecta contradicciones entre los encabezados de archivos locales ZIP, los registros del directorio central y las características de la carga útil. Estas contradicciones pueden utilizarse para...

ai-security-battle

🤖 AI安全对战系统 红队 vs 蓝队：对抗性机器学习战斗模拟 一个全面的对抗性机器学习安全模拟，红队攻击者与蓝队防御者在30分钟AI/ML安全战场景中展开对抗。 🎯 项目概述 本系统模拟真实世界的AI/ML安全场景，其中： 🔴 红队 发起高级攻击（对抗性输入、模型窃取、逃逸攻击） 🔵 蓝队 部署基于ML的防御（异常检测、速率限制、输入验证） 战斗持续30分钟，共10轮，最终结果决定AI/ML安全系统对现代应用是否足够健壮。 📁 项目结构 ai-security-battle/ ├── database/ 数据库层 │ ├── schema.sql 数据库模式 │ ├──...

CVE-2026-79179

A flaw was found in Chromium's Document Object Model DOM due to incorrect authorization. A remote attacker could exploit this vulnerability by crafting a malicious HTML page. This could potentially lead to the leakage of sensitive information...

authproof-sdk

AuthProof SDK AuthProof 是一种面向代理型 AI 的加密授权协议。该领域的大多数协议都基于操作员定义的策略执行——赋予操作员在事后扩展或重新解释用户原始意图的权力。AuthProof 基于不同的信任模型构建：用户自己的私钥签署控制执行的授权对象，并且在授权时和执行前立即验证实时模型状态。用户签署的权限与实时模型状态门控的结合是具体的主张——而非广泛的执行叙事。 不同之处在于： 用户是签署权威。 每个竞争协议（AIP、AITH、OAP、SAGA、AgentSpec）都基于操作员定义的策略执行。在 AuthProof...

sentinel-mcp

SentinelMCP — AI 搭載サイバー防御システム Google Cloud Hackathon 提出作品 — 自律的な脅威検出と対処を Google Cloud Agent Builder Vertex AI Agent が調整し、Gemini 2.0 Flash、Model Context Protocol MCP、Dynatrace 可観測性を活用。 問題点 セキュリティチームは、一般的なインシデントに対して 30 分の MTTD と 4 時間の MTTR...

OWASP-Model-Card-Security-Standard

OWASP 模型卡安全标准 状态：草案 — OWASP 孵化项目 本标准正在积极开发中。非常欢迎贡献和反馈。 这是什么？ 当你从 HuggingFace 或任何其他仓库下载预训练模型时，你本质上是在信任一个黑盒。你不知道它是在什么数据上训练的，这些数据是否包含个人信息，是否做过对抗性攻击测试，也不知道它已知的失败模式是什么。对大多数业余项目来说，这没问题。但对于任何涉及真实用户、受监管行业或敏感数据的场景——这就真的不行了。 模型卡已经存在了一段时间，而且是个好主意。但现有的格式几乎完全集中在性能指标和公平性考量上。安全在很大程度上是事后才想到的，甚至可能根本没有涉及。...