Critical Alert

CVE-2026-50751 - Critical Check Point VPN Exploit Discovered Active in the Wild. View Threat Details →

Powered by CVE WATCHTOWER

×

Prompt injection

Toxic Agent Flow: GitHub MCP Vulnerability Exposes Private Repositories GitHub MCP vulnerability AI agent hijacking

GitHub MCP vulnerability AI agent hijacking

Toxic Agent Flow: GitHub MCP Vulnerability Exposes Private Repositories

Do Son May 28, 2025

Invariant Labs has revealed a critical vulnerability in the widely used GitHub MCP server that enables attackers...

Critical Flaws in AI Browse Agents: Exposed to Credential Theft and Hijacking Prompt injection, agent hijacking

Prompt injection, agent hijacking

Critical Flaws in AI Browse Agents: Exposed to Credential Theft and Hijacking

Do Son May 22, 2025

As AI-powered browsing agents increasingly automate complex web tasks—from booking travel to managing emails—they’re becoming both indispensable...

AI Jailbreaks Expose Systemic Vulnerability in Generative AI Platforms AI Jailbreak Generative AI

AI Jailbreak Generative AI

AI Jailbreaks Expose Systemic Vulnerability in Generative AI Platforms

Do Son April 25, 2025

A newly disclosed vulnerability note from the CERT Coordination Center (CERT/CC) has shed light on two systemic...

Google’s Agentic AI Security Team Develops Framework to Combat Prompt Injection Attacks Gemini AI Prompt injection

Gemini AI Prompt injection

Google’s Agentic AI Security Team Develops Framework to Combat Prompt Injection Attacks

Do Son January 29, 2025

Google’s Agentic AI Security Team announced in a recent blog post that they have developed a new...

ChatGPT’s New Feature Allows Hackers to Steal Your Data prompt injection attack

prompt injection attack

ChatGPT’s New Feature Allows Hackers to Steal Your Data

Do Son November 15, 2023

Users of the premium service ChatGPT Plus can now benefit from an integrated Python interpreter, which simplifies...