agent-framework-azure-ai-py

You can build Azure AI Foundry agents using the Microsoft Agent Framework Python SDK by leveraging the AzureAIAgentsProvider to create persistent agents, integrate hosted tools, and manage conversation threads. Run `pip install agent-framework --pre` to start.

Yes, the Microsoft Agent Framework supports integrating Model Context Protocol (MCP) servers with Azure AI Foundry agents, allowing you to extend agent functionality and connect to additional external tools and data sources.

Yes, you can utilize hosted tools including code interpreter, file search, and web search with Azure AI Foundry agents. The Microsoft Agent Framework Python SDK enables these extended capabilities for your persistent agents.

The best way to manage long-running conversations is by creating persistent agents with AzureAIAgentsProvider. This approach within the Microsoft Agent Framework Python SDK handles conversation threads and maintains state effectively.

Yes, developing persistent agents with the Microsoft Agent Framework requires both the `agent-framework` and `azure-ai-agents` dependencies, alongside Python and the AzureAIAgentsProvider SDK to function correctly.

Using AzureAIAgentsProvider requires a Python environment and specific SDK dependencies. While it supports hosted tools and MCP integration, your agent development is constrained by the available hosted tools and the Azure AI Foundry platform capabilities.