Agent Framework Governance
Policies, standards, and best practices for Microsoft Agent Framework development
The Microsoft Agent Framework provides a comprehensive platform for building, deploying, and managing AI agents using Semantic Kernel, Azure AI services, and industry best practices.
Semantic Kernel Integration
requiredBuild agents using Microsoft Semantic Kernel for function calling, prompt management, and LLM orchestration
Security by Design
requiredImplement robust authentication, authorization, and data protection measures from the start
Responsible AI
requiredFollow Responsible AI principles including transparency, fairness, and accountability
Observability
recommendedImplement comprehensive logging, monitoring, and tracing for all agent interactions
Security & Compliance
Development Standards
Single Agent Pattern
Simple assistant for focused tasks with defined scope
Orchestrator Pattern
Coordinator agent that delegates to specialized sub-agents
Multi-Agent System
Collaborative agents working together on complex problems