How to Create AI Agents Like a Pro in Copilot Studio (No Code Required) by Microsoft Press

Course details

Microsoft Copilot Studio contributes to the democratization of AI automation. This course is designed to help transform business users into automation architects. Learn how to build intelligent agents that handle everything from customer inquiries to document processing—all without writing a single line of code.

Learning objectives
Build conversational and autonomous agents using natural language in Copilot Studio.
Connect agents to SharePoint, OneDrive, Dataverse, and external knowledge sources.
Configure autonomous triggers to wait for specific events and execute actions automatically.
Implement agent flows for repeatable business processes that require consistency and control.
Enable deep reasoning and GPT-5-powered agents for advanced multi-step analysis.
Deploy agents to Teams, SharePoint, Microsoft 365 Copilot, and the Agent Store marketplace with 70+ prebuilt agents.
Measure ROI using new analytics features that track estimated time and cost savings per action.

Concepts

Introduction

How to create AI agents like a pro in Copilot Studio - Introduction

The Agent Confusion Crisis

Learning objectives
Microsoft s agents versus industry agents - What s real
Autonomous agents - What you can build today
Agent templates, managed agents, and the Agent Store
Your first Hello World agent in 5 minutes
Demo - Prompting in M365 Copilot
Demo - No-code agents in M365 Copilot Agent Builder
Demo - Low-code agents in Microsoft Copilot Studio
Demo - Code-first agents in Microsoft Foundry

Agent #1 - FAQ Responder

Learning objectives
Natural language agent creation using Describe tab
Add SharePoint knowledge sources
Configure instructions and conversation style
Build and ship your first production agent
Demo - Ground your agent in organizational data
Demo - Building out a topic structure
Demo - Integrating tool-calling into our agent s behavior
Demo - Introducing the Copilot Studio extension for Visual Studio Code
Demo - Publishing your agent to M365 Copilot and Teams

Power Automate Foundation

Learning objectives
Understand actions versus topics versus flows
Agent flows vs. cloud flows
The agent flow pattern
Variables and decision logic without code
Demo - Deeper into Copilot Studio topic flows
Demo - Designing and building an agent flow
Demo - Debugging the agent flow

Agent #2 - IT Support Ticket System

Learning objectives
From chatbot to business system
Ticket creation and approval workflows
Multi-step conversations and topic routing
Testing and pitfall avoidance
Demo - Review the agent s structure
Demo - Experimenting with conversational boosting
Demo - Testing the agent
Demo - More work with agent flows

Autonomous and Deep Reasoning Capabilities

Learning objectives
From reactive to autonomous
Event triggers and payloads
AI models and deep reasoning
Demo - Sharpening our agent s custom instructions
Demo - Implementing tools and agent flows
Demo - Debugging the agent

Agent #3 - Document Processing Automation

Learning objectives
Unstructured data and the dark matter problem
Knowledge indexing and semantic search
Deterministic processing and notifications
Demo - Reviewing the agent s runtime behavior
Demo - Testing and optimizing the agent

Deployment and Analytics

Learning objectives
Building vs. deploying
Teams and the organizational catalog
Copilot credits and ROI
Analytics and governance
Demo - Revisiting agent deployment
Demo - Governing your Copilot Studio agents

Building Your Agent Roadmap

Learning objectives
Discovery and opportunity matrix
Discovery and prioritization
Phased roadmap and next steps
Certifications and community
Demo - Analyzing Copilot billing and licensing
Demo - Touring the Microsoft Agent Usage Estimator

Summary

How to create AI agents like a pro in Copilot Studio - Summary