← Back to Tutorials

13. References

Glossary

TermDefinition
Foundry HubShared resource container (AI Search, Storage, Key Vault)
Foundry ProjectWorkspace within a Hub for building AI applications
Agent ServiceManaged runtime for AI agents with tools and memory
RAGRetrieval-Augmented Generation — ground LLM responses in retrieved data
MCPModel Context Protocol — standardized tool interface
A2AAgent-to-Agent protocol for inter-agent communication
AI GatewayAPI Management layer for AI models and agents
Foundry IQAgentic retrieval with intelligent ranking

Cheatsheet

# Python SDK Quick Reference

# 1. Create project client
from azure.ai.projects import AIProjectClient
client = AIProjectClient(subscription_id, resource_group, project_name, credential)

# 2. Create agent
agent = client.agents.create_agent(model="gpt-4o", name="MyAgent", instructions="...")

# 3. Add tools
from azure.ai.projects.models import ToolSet, CodeInterpreterTool, WebSearchTool
toolset = ToolSet()
toolset.add(CodeInterpreterTool())

# 4. Send message
response = client.agents.send_message(agent.id, "Hello")

# 5. Create search index
from azure.search.documents.indexes import SearchIndexClient
index_client = SearchIndexClient(endpoint, credential)
index_client.create_index(index_definition)

# 6. Deploy model (portal or CLI)
# az ml model create --name gpt-4o --type deployment

API Reference

Continue Your Journey

📘 Congratulations! You've completed the Microsoft Foundry Fundamentals guide. You now have the knowledge to build, deploy, and manage AI agents at scale using Azure AI Foundry.