Copilot Chat, agent mode and metrics

GH-300 · GitHub Copilot plans and features · Updated

Where Copilot Chat runs, how ask, edit and agent mode differ, and how admins measure adoption with usage metrics, seat APIs and audit logs.

This lesson is part of full GH-300 access

The first 2 GH-300 lessons are free. Full access opens every lesson with its practice questions, and the exam-style GH-300 question bank.

In this lesson
  • Ways to interact
  • Chat modes
  • Model choice and custom models
  • Measuring usage
  • Policies and governance
  • Exam traps
  • Practice questions

Get full access Try the free practice exam Log in

GH-300 course outline

All 7 lessons · Course overview
  1. Responsible AI principles for CopilotResponsible AI Free
  2. GitHub Copilot plans comparedGitHub Copilot plans and features Free
  3. Copilot Chat, agent mode and metricsGitHub Copilot plans and features Full access
  4. How Copilot builds a response and handles dataHow GitHub Copilot works and handles data Full access
  5. Prompt crafting for CopilotPrompt crafting and prompt engineering Full access
  6. Developer use cases across the SDLCDeveloper use cases Full access
  7. Content exclusions and troubleshootingTesting, privacy and context exclusions Full access