Collected every two hours from specialised publications — each link leads to the original article.
A flaw in the OpenCode AI coding agent allowed a malicious or hijacked MCP server to run code on a developer's Windows or WSL machine when the developer signed…
In 2024, MCP (Model Context Protocol) set out to become the USB-C of AI: one standard for connecting models, agents, and IDEs to tools and data. The protocol d…
A malicious MCP server could trick an application built on the official MCP Python SDK into handing over the OAuth credentials it uses to log in to a real serv…
A practical checklist for the Azure DevOps MCP flaw that lets hidden PR comments hijack AI coding agents, plus the configuration changes to make right now.Azur…
CVE-2026-12957 in Amazon Q is the third MCP auto-execution vulnerability in three AI coding tools. The pattern reveals a shared design failure, not just a sing…
Cybersecurity researchers have discovered a critical "by design" weakness in the Model Context Protocol's (MCP) architecture that could pave the way for remote…
Cybersecurity researchers have disclosed details of a now-patched vulnerability in the popular figma-developer-mcp Model Context Protocol (MCP) server that cou…
Cybersecurity researchers have disclosed details of a now-patched vulnerability in the popular figma-developer-mcp Model Context Protocol (MCP) server that cou…
Cybersecurity researchers have discovered a critical vulnerability in the open-source mcp-remote project that could result in the execution of arbitrary operat…
An MCP security flaw in Google's open-source Toolbox for Databases could let an attacker copy files off the server running it or replace files on that machine.
A September 23 advisory describes a flaw in the Python SDK used with MCP Toolbox: a shared cache could send a Google ID token to a service it was not meant for.
An update for mcphost-0.34.0-9.1 on openSUSE Tumbleweed addresses two security vulnerabilities with moderate ratings, allowing installation to enhance system s…
A high-severity flaw in Amazon Q Developer let a malicious repository run commands and steal a developer's cloud credentials. The path was short: a developer o…
Cybersecurity researchers have disclosed a high-severity security flaw in the artificial intelligence (AI)-powered code editor Cursor that could result in remo…
Cybersecurity researchers have discovered a critical security vulnerability in artificial intelligence (AI) company Anthropic's Model Context Protocol (MCP) In…
libXdmcp could be made to expose sensitive information.
An update that solves 3 vulnerabilities and has 3 bug fixes can now be installed.
An update that solves two vulnerabilities can now be installed.
An update that solves 20 vulnerabilities and has 4 bug fixes can now be installed.
An update that solves 3 vulnerabilities can now be installed.
An update that solves 3 vulnerabilities can now be installed.
A recently disclosed critical security flaw impacting nginx-ui, an open-source, web-based Nginx management tool, has come under active exploitation in the wild…
Vulnerability in the strided image data parsing code in the emscripten wrapper for libheif. An attacker could exploit this through a crafted image file to caus…
Null source pointer passed as an argument to memcpy() function within TIFFFetchStripThing() in tif_dirread.c in libtiff versions from 3.9.0 to 4.3.0 could lead…