AI Security
Lessons from the field. Always landing on my feet.
Agent 365 Launch Playbook: I Tested the Defender Response for AI Agent Attacks
When I published this lab on April 20, Microsoft had announced that Agent 365 would become generally available on May 1, 2026. It reached that milestone; the launch-week security question behind the lab still matters. I wanted to answer a different …
Block Prompt Injection at the Network Layer with Microsoft Entra Prompt Injection Protection
A while back I built an LLM Firewall with AWS Lambda, a proxy that sits between users and the model to catch prompt injection. It worked, but it meant writing custom code for every app and having zero visibility into AI services I didn’t own. That’s …
Just-In-Time Access for AI Agents: Building a ZSP Gateway in Azure
AI coding assistants need Contributor access to deploy infrastructure. Backup automation needs Key Vault secrets at 2 AM. Security scanners need Reader access on a schedule. The easy answer is standing permissions—give each service principal what it …
Building an LLM Prompt Injection Firewall with AWS Lambda
AWS continues to enhance its generative AI security capabilities, with improved prompt attack filtering now available in Amazon Bedrock Guardrails. Despite these advances, a significant gap remains: organizations are deploying LLM capabilities faster …
Sentinel MCP Server: Securing Your SOC's New AI Attack Surface
In September 2025, Microsoft announced the Sentinel MCP Server, a Model Context Protocol implementation that lets MCP-compatible AI assistants query your Sentinel data using natural language. Microsoft highlights GitHub Copilot, Copilot Studio, and …
Securing the Agentic Workforce: Microsoft's Zero Trust for AI Agents
Status update — August 13, 2026: Microsoft Agent 365 is generally available for commercial customers, effective May 1, 2026, and the Microsoft Entra Agent ID platform reached GA in April 2026. The Ignite 2025 rollout history below remains useful …