Copilot Chat is powerful on its own, but some tasks require more than a conversation. This module introduces the distinction between general chat and specialized agent-based experiences, giving learners the analytical framework to identify when a task warrants an agent and how to recognize the difference in practice.

Microsoft Graph is not limited to Microsoft 365 data. This module introduces Graph connectors, the mechanism that allows Copilot to draw from external organizational data sources like Salesforce, ServiceNow, or internal databases, giving learners a complete picture of how Copilot's Work mode can be extended across the full data ecosystem of a modern organization.

This module provides learners with a structured, repeatable method for writing prompts that yield professional-grade outputs on the first attempt. The Clarity, Context, and Constraints framework replaces guesswork with intention and reframes technical-skill prompts as something every professional already knows how to do: giving clear instructions to a capable colleague.

Even a well-structured prompt may not yield a finished product on the first attempt. This module teaches learners to treat Copilot as a collaborative drafting partner, using targeted follow-up prompts to refine tone, restructure content, and verify factual accuracy through a deliberate iterative process. Think of it as giving feedback to a capable intern: not starting over, but redirecting with precision.

Before learners can orchestrate agents effectively, they need to understand the landscape. This module introduces the two paths to agent capability—the Agent Store for pre-built solutions and Microsoft Copilot Studio for custom builds—and provides learners with a clear framework for deciding which path best fits a given business need.

This module moves from understanding agents to directing them. Learners will apply a task analysis framework to select between the Researcher and Analyst agents, then practice invoking and grounding agents using the @mention syntax across real business scenarios, including deep market research and complex multi-source data analysis.

Pre-built agents cover a lot of ground, but some business needs require a custom solution. This module introduces Microsoft Copilot Studio as a platform for building agents tailored to specific organizational workflows, giving learners enough understanding of the configuration process to evaluate feasibility, brief a technical team, or build a basic agent themselves.

Writing effective prompts is a skill; writing responsible prompts is a professional standard. This module extends the Clarity, Context, and Constraints framework into the territory of data privacy and intellectual property, equipping learners with practical techniques for producing the outputs they need without exposing sensitive information or creating IP risk in client-facing work.

Learners step into the role of a Lead Analyst at Helix Health, a diagnostic hardware company that went all-in on Microsoft 365's Microsoft Copilot six months ago. The tool is capable. The problem is that the team has not yet developed the skills or shared standards to use it effectively. Without a common framework for mode selection, grounding, and prompt structure, every team member is experimenting independently and getting inconsistent results. The shared #AI-Experimentation channel has become a collection of fragmented, underspecified prompts that are not giving Microsoft Copilot enough to work with. Your job is to close that gap. You will audit the team's Fragmented Files, apply the Clarity, Context, and Constraints framework to repair each prompt into a professional Enterprise Prompt Library, and design an Agent Workflow that solves the P1 Diagnostic Sync bottleneck—a process where technicians are manually bridging Microsoft Teams transcripts, technical manuals, and ServiceNow maintenance logs every time a lab sequencer goes offline.