Wiz launches Red Agent for AI vulnerability simulation

Wiz has significantly broadened its AI Application Protection Platform, extending security coverage into AI coding tools, agent studios, multicloud platform services, and internet edge infrastructure while launching its Red Agent in public preview for modeling attacker behavior to validate complex vulnerabilities, reports Security Brief Australia.

The expansion adds support for Databricks and studio environments, including AWS Agentcore, Gemini Enterprise Agent Platform, and Salesforce Agentforce, addressing the risk created when autonomous agents gain access to live data. Complementing the already available Blue Agent and preview-phase Green Agent, Red Agent is designed to identify and exploit weaknesses through offensive simulation.

Wiz also introduced the Technology Intel Centre to track feature releases and end-of-life notices across cloud and AI providers. New code-level capabilities include AI-BOM inventorying of frameworks like LangChain and security guardrails that scan AI-generated code in tools such as Lovable before commit. Pre-built remediation skills for environments like Claude Code and Cursor allow AI agents to pull validated findings directly into developer consoles. Edge integrations with Cloudflare, Akamai, Vercel, and Apigee aim to give security teams visibility from the first AI-generated line of code through to the cloud perimeter.