Cybersecurity researchers have discovered a critical "by design" weakness in the Model Context Protocol's (MCP) architecture ...

CVE-2026-5760 (CVSS 9.8) exposes SGLang via /v1/rerank endpoint, enabling RCE through malicious GGUF models, risking server ...

The Ruby vulnerability is not easy to exploit, but allows an attacker to read sensitive data, start code, and install ...

Remote terminal units, PLCs, PoS systems, and bedside patient monitors may be susceptible to remote code execution, ...

Administrators of websites running the Drupal content management software (CMS) are urged to take immediate action to mitigate a newly discovered a vulnerability that can lead to remote execution of ...

A prompt injection flaw in Google’s Antigravity IDE turns a file search tool into a remote code execution vector, bypassing ...

Join Push Security's free three-part webinar series where top security researchers break down the browser-based attacks driving today's biggest breaches - from AiTM phishing and ConsentFix to ...

The post OpenClaw-Based AI Agents Exposing 28,000 Systems to Hackers, Research Finds appeared first on Android Headlines.

A design choice in the MCP SDKs allows remote code execution across the AI supply chain.

Splunk has released patches that resolve high- and medium-severity vulnerabilities in Splunk Enterprise and MCP Server.

Microsoft releases an out-of-band hotpatch for critical Windows 11 RRAS vulnerabilities that could allow remote code execution through malicious remote servers. Google released a Chrome security ...

A surge in AI agent adoption is exposing critical systems online ...