NPM: Nuxt: `__nuxt_island` endpoint does not bind responses to request props, enabling shared-cache poisoning

NPM: Nuxt: nuxtisland endpoint does not bind responses to request props, enabling shared-cache poisoning vulnerability discovered by ? in WordPress Npm nuxt versions = 3.1.0, = 3.21.5...

HTTP Request Smuggling

Overview Affected versions of this package are vulnerable to HTTP Request Smuggling via the nuxtisland endpoint when responses are not properly bound to request props, allowing shared-cache poisoning. An attacker can cause users to receive attacker-controlled HTML by priming a shared cache with...

CVE-2026-47107

Windmill prior to 1.703.2 contains an incorrect default permissions vulnerability in nsjail sandbox configuration files where /etc is bind-mounted without read-write restrictions, allowing authenticated users to write arbitrary entries to /etc/hosts, /etc/resolv.conf, and...

EUVD-2026-30958

Windmill prior to 1.703.2 contains an incorrect default permissions vulnerability in nsjail sandbox configuration files where /etc is bind-mounted without read-write restrictions, allowing authenticated users to write arbitrary entries to /etc/hosts, /etc/resolv.conf, and...

dnsmasq: dnsmasq: heap buffer overflow in cache via NAME_ESCAPE expansion

A heap buffer overflow was discovered in dnsmasq's DNS cache. When processing DNS responses, dnsmasq expands certain characters into longer escape sequences, but the cache buffer is not sized to hold the expanded result. A specially crafted DNS response can overflow this buffer, potentially...

node-tar: tar: node-tar: Arbitrary file overwrite and symlink poisoning via unsanitized linkpaths in archives

A flaw was found in the node-tar library. This vulnerability allows an attacker to craft malicious archives that, when extracted, can bypass intended security restrictions. This leads to arbitrary file overwrite and symlink poisoning, potentially allowing unauthorized modification of files on the...

CVE-2026-43634 HestiaCP 1.2.0-1.9.4 IP Spoofing via CF-Connecting-IP Header

HestiaCP versions 1.2.0 through 1.9.4 contain an IP spoofing vulnerability that allows unauthenticated remote attackers to bypass authentication security controls by supplying an arbitrary IP address in the CF-Connecting-IP HTTP header without verifying the request originated from Cloudflare's...

node-tar: tar: node-tar: Arbitrary file overwrite and symlink poisoning via unsanitized linkpaths in archives

A flaw was found in the node-tar library. This vulnerability allows an attacker to craft malicious archives that, when extracted, can bypass intended security restrictions. This leads to arbitrary file overwrite and symlink poisoning, potentially allowing unauthorized modification of files on the...

Compromised Nx Console 18.95.0 Targeted VS Code Developers with Credential Stealer

Cybersecurity researchers have flagged a compromised version of the Nx Console extension that was published to the Microsoft Visual Studio Code VS Code Marketplace. The extension in question is rwl.angular-console version 18.95.0, a popular user interface and plugin for code editors like VS Code,...

PT-2026-41762

AutoGPT is a workflow automation platform for creating, deploying, and managing continuous artificial intelligence agents. In versions 0.6.34 through 0.6.51, the backend deserializes Redis cache bytes using pickle.loads without integrity/authenticity checks. The write path serializes values with...

RHEL 10 : linux-sgx (RHSA-2026:18480)

The remote Redhat Enterprise Linux 10 host has packages installed that are affected by multiple vulnerabilities as referenced in the RHSA-2026:18480 advisory. The Intel SGX SDK is a collection of APIs, libraries, documentations and tools that allow software developers to create and debug Intel SG...

RHEL 9 : linux-sgx (RHSA-2026:18868)

The remote Redhat Enterprise Linux 9 host has packages installed that are affected by multiple vulnerabilities as referenced in the RHSA-2026:18868 advisory. The Intel SGX SDK is a collection of APIs, libraries, documentations and tools that allow software developers to create and debug Intel SGX...

Important: linux-sgx security update

The Intel SGX SDK is a collection of APIs, libraries, documentations and tools that allow software developers to create and debug Intel SGX enabled applications in C/C++. Security Fixes: qs: qs: Denial of Service via improper input validation in array parsing CVE-2025-15284 node-tar: tar: node-ta...

FreeBSD : Vinyl/Varnish -- HTTP/2 parsing deficiency (f0f4bb64-52c6-11f1-a1c0-0050569f0b83)

The version of FreeBSD installed on the remote host is prior to tested version. It is, therefore, affected by a vulnerability as referenced in the f0f4bb64-52c6-11f1-a1c0-0050569f0b83 advisory. Vinyl Development Team reports: A deficiency in HTTP/2 request parsing can be exploited to launch a...

Windmill 安全漏洞

Windmill is a low-code development platform open source by Windmill Labs, Inc. Versions of Windmill prior to 1.703.2 contained security vulnerabilities. These vulnerabilities stemmed from the binding and mounting of directories under /etc in the nsjail sandbox configuration file without any...

PT-2026-42038

Name of the Vulnerable Software and Affected Versions Nuxt versions 3.1.0 through 3.21.5 Nuxt versions 4.0.0-alpha.1 through 4.4.5 @nuxt/nitro-server versions 3.20.0 through 3.21.5 @nuxt/nitro-server versions 4.0.0-alpha.1 through 4.4.5 Description The '/ nuxt island/' endpoint accepts...

ALSA-2026:18868 Important: linux-sgx security update

The Intel SGX SDK is a collection of APIs, libraries, documentations and tools that allow software developers to create and debug Intel SGX enabled applications in C/C++. Security Fixes: qs: qs: Denial of Service via improper input validation in array parsing CVE-2025-15284 node-tar: tar: node-ta...

Exploit for Embedded Malicious Code in Tanstack Tanstack\/Arktype-Adapter

TanStack Supply Chain Compromise - IOC Checker bash curl -...

Be Kind, Rewrite: Benign Projections Via Rewriting Defend against LLM Data Poisoning Attacks

Large language models LLMs are highly susceptible to backdoor attacks BAs, wherein training samples are poisoned using trigger-based harmful content. Furthermore, existing defenses have proven ineffective when extensively tested across BA patterns. To better combat BAs, we explore the use of LLM...

Token by Token, Compromised: Backdoor Vulnerabilities in Unified Autoregressive Models

Unified autoregressive models UAMs are transformer models that generate text as well as image tokens within a single autoregressive pass. Shared parameters and a multimodal vocabulary simplify the training pipeline and facilitate flexible multimodal generation, yet might introduce new...