Design and configure agents in Microsoft Copilot Studio (GitHub Copilot)

Plan and configure agents powered by the GitHub Copilot harness in Microsoft Copilot Studio. Manage usage and cost, define agent behavior with instructions, and package reusable behavior as skills.

Prerequisites

  • Basic understanding of artificial intelligence and agents.
  • Access to Microsoft Copilot Studio.
  • Basic familiarity with Markdown syntax.

Modules in this learning path

Microsoft Copilot Studio supports agents and workflows with different runtime architectures. This module explains how harness choice affects orchestration, available capabilities, authoring, publishing, and billing. You compare the Microsoft Copilot, Customer Conversation, and standard harnesses and select a harness for a business scenario.

Manage usage and cost for agents in Microsoft Copilot Studio. Compare how each agent harness meters Copilot Credits, identify the design and build-time decisions that drive consumption on the GitHub Copilot harness, and describe how administrators fund and monitor capacity.

Agents powered by the Microsoft Copilot harness use natural-language instructions as the primary way to define their behavior. This module explains how to define an agent's purpose, scope, behavior, boundaries, and grounding, and then refine the agent by testing it in Preview.

Package reusable behavior as skills for agents powered by the GitHub Copilot harness in Microsoft Copilot Studio. Decide when a skill is the right component, author and package skills, and reference them from agent instructions across agents and environments.