AI Agents - 3

2026 (3)

June (2)

ai: Build agents you can trust across any framework with open evals and a control standard

Microsoft AI Foundry Blog

2026-06-02

Microsoft describes a set of tools for building AI agents with open evaluations, portable runtime controls, observability, and security workflows that are intended to work across frameworks. The post focuses on trust and control for agent development rather than a single framework-specific feature set.

ai: What’s New in Hosted Agents in Foundry Agent Service

Microsoft AI Foundry Blog

2026-06-02

Microsoft announced new updates for hosted agents in Foundry Agent Service at Build, focused on easier deployment, broader modality support, and tighter integration for enterprise use. The article is an overview of recent hosted agent runtime improvements.

May (1)

security: Introducing RAMPART and Clarity: Open source tools to bring safety into Agent development workflow

Microsoft Security Blog

2026-05-20

Microsoft announced two open-source tools, RAMPART and Clarity, aimed at adding safety checks to agent development workflows. The post frames the need around enterprise AI agents that can access data, execute code, and take actions across connected systems.