Infosecurity Europe: AI Coding Tools Need Built-In Security for Agentic Development Era

The article warns that agentic AI coding agents create new, fast-moving attack surfaces (input, tools, execution, output) that traditional appsec cannot handle and advocates embedding continuous, autonomous security checks into the development loop; it also highlights a critical May 2026 vulnerability in the Cline Kanban server that could enable hijacking of AI coding tools.