LaunchKit · 2026
Back to Skills

azure-ai-agents-py

Build AI agents using the Azure AI Agents Python SDK.

0
1.8k downloads
by @thegovind

Setup & Installation

openclaw skills install @thegovind/azure-ai-agents-py

Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:

npx clawhub install azure-ai-agents-py

What This Skill Does

Python SDK for building agents hosted on Azure AI Foundry. Handles the full agent lifecycle: creating agents, managing conversation threads, processing runs, and streaming responses. Supports tools including Code Interpreter, File Search, Bing Grounding, Azure AI Search, Function Calling, OpenAPI, and MCP.

Gives direct SDK access to Azure AI Foundry's agent infrastructure with full control over thread management, tool configuration, and streaming, without higher-level framework abstractions.

When to use it

  • Building a document Q&A agent over uploaded PDFs using File Search
  • Running on-demand Python code execution via Code Interpreter
  • Grounding agent responses with live web search via Bing
  • Streaming real-time agent replies in a chat interface
  • Calling internal REST APIs from an agent using the OpenAPI tool

Example Workflow

Here's how your AI assistant might use this skill in practice.

INPUT

User asks: 'What is the weather in Seattle?'

AGENT
  1. 1Create an agent with a FunctionTool wrapping a get_weather Python function
  2. 2Create a conversation thread and add the user message
  3. 3Call create_and_process with the toolset to auto-execute the function
  4. 4Retrieve completed run assistant messages from the thread
  5. 5Return the weather response text
OUTPUT

Weather in Seattle: 58F, cloudy

Requirements

Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.

Azure subscriptionAzure AI Foundry project with a deployed modelPROJECT_ENDPOINT environment variable (from Azure AI Foundry project settings)MODEL_DEPLOYMENT_NAME environment variable (name of your deployed model)