Israeli cybersecurity firm Vega launches open AI threat detection standard

Israeli cybersecurity firm Vega has launched Detection Skills, an open standard designed to enhance threat detection and investigation for security teams. The framework, publicly available via DetectionSkills.io and GitHub, allows for the capture, sharing, and automation of expert reasoning in AI-powered workflows, addressing the challenge of rapidly evolving AI-driven cyber threats. Vega co-founder and CTO Eli Rozen stated, "AI-driven adversaries bypass static rules in every legacy SIEM, and no rule catches an attack it has never seen." The standard aims to provide a common language for AI-assisted detection, similar to Sigma for security detection rules, and includes an Agentic Detection Library with over 35 pre-developed skills. It integrates with existing cybersecurity infrastructure without requiring system replacements, with Peloton and Exemplar Luxury Group already utilizing the framework to improve alert triage and maintain operations during peak activity.