Can I use custom Python functions as tools?

Yes, you can pass standard Python functions directly to the agent; the framework automatically converts them into AIFunction tools using your type hints and docstrings.

Does it support streaming responses?

Yes, it includes patterns for using agent.run_stream() to provide real-time, token-by-token output for a better user experience.

Which hosted tools are supported?

The skill supports Hosted Code Interpreter for Python execution, Hosted Web Search via Bing, and Hosted File Search for querying vector stores.

How does this skill handle conversation history?

It uses the AgentThread class to manage conversation persistence, allowing agents to maintain context across multiple turns and sessions hosted on Azure.

What is the Microsoft Agent Framework Python SDK?

It is a specialized toolkit designed to build, run, and scale AI agents on Azure AI Foundry, offering native support for persistent threads and enterprise-grade tool integrations.

Azure AI Agent Framework (Python)

Name: Azure AI Agent Framework (Python)
Author: sickn33

bysickn33

•

31,722

•

Infraestructura en la Nube

Builds and manages persistent, multi-tool AI agents on Azure AI Foundry using the Microsoft Agent Framework Python SDK.

This skill provides a comprehensive toolkit for developing and deploying stateful, production-grade AI agents within the Azure ecosystem. It streamlines the implementation of complex agentic workflows, including persistent conversation threads, structured outputs, and real-time streaming. By leveraging the Microsoft Agent Framework, developers can easily integrate advanced capabilities such as hosted code execution, Bing web search, and Model Context Protocol (MCP) tools. This skill is ideal for building enterprise-ready assistants that require high reliability, deep Azure integration, and robust tool-calling capabilities.

Características Principales

01Advanced response handling including streaming and validated structured outputs

02Persistent conversation management via Azure-hosted AgentThreads

03Support for both service-managed and client-managed MCP tools

0431,722 GitHub stars

05Simplified function tool registration with Pydantic and type-hinting

06Seamless integration with Hosted Code Interpreter and Bing Web Search

Casos de Uso

01Creating stateful customer support agents that maintain context across long sessions

02Building enterprise research assistants with web search and data analysis capabilities

03Developing automated data processing pipelines that require validated JSON outputs

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add sickn33/antigravity-awesome-skills agent-framework-azure-ai-py

For use in Claude.ai and ChatGPT

Características Principales

01Advanced response handling including streaming and validated structured outputs

02Persistent conversation management via Azure-hosted AgentThreads

03Support for both service-managed and client-managed MCP tools

0431,722 GitHub stars

05Simplified function tool registration with Pydantic and type-hinting

06Seamless integration with Hosted Code Interpreter and Bing Web Search

Casos de Uso

01Creating stateful customer support agents that maintain context across long sessions

02Building enterprise research assistants with web search and data analysis capabilities

03Developing automated data processing pipelines that require validated JSON outputs

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add sickn33/antigravity-awesome-skills agent-framework-azure-ai-py

For use in Claude.ai and ChatGPT