7 articles tagged with "Mcp"
Microsoft discloses AutoJack, an exploit chain that hijacks AutoGen Studio AI agents via malicious webpages. A single URL visit triggers arbitrary code execution on the host machine.
Researchers at Tenet Security discovered Agentjacking, an attack that tricks AI coding assistants like Claude Code and Cursor into executing arbitrary code through malicious Sentry error events.
CVE-2026-40933 (CVSS 9.9) allows attackers to compromise self-hosted Flowise AI agent builders by tricking users into importing a malicious chatflow. The payload executes during import without user action.
CVE-2026-33032 lets attackers take full control of nginx-ui servers without credentials. Threat actors are exploiting it now. Upgrade to 2.3.4 immediately.
Cisco's State of AI Security 2026 report reveals a dangerous gap between agentic AI adoption ambitions and enterprise security readiness. Here's what the threat landscape looks like.
New catalog at developer.cisco.com/codeexchange/ai centralizes AI agents and MCP servers for network automation, with built-in testing tools.
From AI-driven network automation to Meraki's OAuth 2.0 rollout, here's what developers should target at Cisco Live Amsterdam.