// Track

GitHub Copilot CLI

The AI Agent in Your Terminal

GitHub's terminal-native AI agent — install it, master the session workflow, wire custom instructions, use plan mode for complex tasks, delegate async work, and set up your team for production use.

8 lessons~64 min total

Lessons are shown in recommended order. Complete them in sequence for the best experience — or jump to any lesson.

Lesson 312·8 min read

What Is GitHub Copilot CLI

GitHub Copilot CLI is not the autocomplete you already know — it is a terminal-native agent that reads your entire project, runs shell commands, and executes multi-step tasks from a single prompt.

Lesson 313·8 min read

Your First Copilot CLI Session

Getting from zero to a running Copilot CLI session takes under five minutes — here is exactly what to install, how to authenticate, and what to do in the first session so you learn the rhythm before you need it for real work.

Lesson 314·8 min read

Custom Instructions

Custom instructions are how you turn a generic AI agent into one that knows your stack, your conventions, and your rules — and the file you check into .github/ is the one that matters most.

Lesson 315·8 min read

Plan Mode

Plan mode is the mechanism that separates confident multi-file changes from hope — Copilot CLI generates an explicit, reviewable plan before touching anything, and you approve it before execution starts.

Lesson 316·8 min read

Session State and Context Management

Copilot CLI accumulates context through tool calls and persists it to disk — understanding what is stored, where, and how to inspect it is the difference between debugging a confused session and catching a problem before it compounds.

Lesson 317·8 min read

Tool Control — Allow and Deny

Copilot CLI's default tool access is broad — configuring explicit allow and deny rules is not paranoia, it is the difference between a useful agent and one that can accidentally push to main or delete data.

Lesson 318·8 min read

Delegating Work with /delegate

/delegate and /fleet let you hand off well-defined subtasks to background agents while your session continues — parallel AI work on a single codebase without the coordination overhead.

Lesson 319·8 min read

Team Setup and Repository Conventions

Getting your team running on Copilot CLI consistently is a configuration problem, not a training problem — the right repo setup means every developer's first session is already calibrated to your standards.