https://catalogartifact.azureedge.net/publicartifacts/insightsoftwareconsulting1729066469769.m365copilotworkshop-a9284892-554d-4138-b07f-bec00da3af31/image4_InsightLogo216.png

Microsoft 365 Copilot Agent - 1 Day Implementation Workshop

INSIGHT SOFTWARE CONSULTING

This workshop introduces participants to building Copilot agents in a single day. It covers the Copilot ecosystem, agent types, Agent Builder, Copilot Studio, data grounding, actions, and governance.

Introduction to the Microsoft 365 Copilot Ecosystem Microsoft's AI vision and the broader Copilot family. How Microsoft 365 Copilot, Microsoft 365 Copilot Chat, Copilot Studio, and Agent Builder relate to one another and where each fits. Key concepts: grounding, Microsoft Graph, foundation models, and the Copilot orchestrator. Licensing models and tenant-level prerequisites. What Is a Copilot Agent? The distinction between declarative agents and custom engine agents with practical use cases. The anatomy of an agent: instructions, knowledge sources, actions, triggers, and conversation starters. A brief look at Microsoft's prebuilt out-of-the-box agents such as Researcher, Analyst, and Sales Agent. Getting Started with Agent Builder Accessing Agent Builder from Microsoft 365 Copilot. Building a declarative agent, defining its purpose, writing effective instructions, adding SharePoint sites or files as knowledge, configuring conversation starters, and publishing. Agent Builder governance principles. Hands-On Lab - Building an Agent in Agent Builder Creating an agent grounded on a sample SharePoint document library. Iterating on instructions, testing responses, refining configuration, and sharing the agent with a peer. Common pitfalls such as over-broad instructions and weak grounding. Introduction to Copilot Studio Microsoft Copilot Studio as the low-code authoring environment for more advanced agent scenarios. When Agent Builder is sufficient and when Copilot Studio is the better fit. Interface walkthrough: topics, knowledge, actions, triggers, channels, analytics, and the test pane. Knowledge and Generative Answers in Copilot Studio Grounding agents in enterprise data using SharePoint, Dataverse, public websites, Microsoft 365 Copilot connectors, and uploaded files. The difference between authored topics and generative answers. Best practices for content curation and citation handling. Actions and Connectors Making agents take action using Power Platform connectors, custom connectors, Power Automate flows, and REST APIs. Connecting an agent to Dataverse, Outlook, Teams, and external systems. Authentication considerations. Hands-On Lab - Building an Agent in Copilot Studio Building an agent that combines knowledge sources with an action. Example scenario: an onboarding agent that answers HR policy questions from SharePoint and creates an IT ticket via Power Automate. Testing, debugging, and iterating. Publishing, Channels, and Governance Publishing agents to Microsoft Teams, Microsoft 365 Copilot, websites, and other channels. Governance topics including the Power Platform admin center, environment strategy, and agent lifecycle management. Overview of Custom Engine Agents What custom engine agents are, when they are needed, and what tools are available: Microsoft 365 Agents Toolkit, Microsoft 365 Agents SDK, Azure AI Foundry. A foundation for participants who want to explore this path further after the workshop. Wrap-Up and Next Steps Open Q&A, curated resource list, and a follow-up exercise to build and share an agent within the organization. #Pricing and delivery terms are customized based on customer requirements.

Overblik

https://catalogartifact.azureedge.net/publicartifacts/insightsoftwareconsulting1729066469769.m365copilotworkshop-a9284892-554d-4138-b07f-bec00da3af31/image3_insightm365copilotagent1dayworkshop.png