Microsoft Open-Sources Agent Governance Toolkit to Secure AI Agents
Another open source release from Microsoft:
This is a new tool for protecting autonomous AI agents from hacking, manipulation, and unauthorized actions.
According to the company, it is the world’s first tool that covers all ten agentic AI risks that OWASP documented last year.
The Toolkit consists of several modules. Agent OS works as a policy engine and intercepts every agent action before it is executed. Agent Mesh secures communication between agents. Agent Runtime manages dynamic execution rings. Agent SRE is responsible for general protective mechanisms. Agent Compliance automatically checks compliance with requirements and assigns scores. Agent Marketplace manages the plugin lifecycle. Agent Lightning provides reinforcement learning management.
Works with LangChain, CrewAI, AutoGen, OpenAI Agents SDK, Semantic Kernel, and other frameworks
Supports any stack: Python, TypeScript, Go, Rust, .NET.