GH-300 cheat sheet
A one-page reference for the GitHub Copilot exam: the format, how the domains are weighted, and the glossary terms for this exam.
Exam at a glance
Vendor
GitHub
Level
Intermediate
Questions
60
Time
90 min
Mock pass mark
70%
Domains
6
Practice Qs
190
Code
GH-300
Domain weightings
How much of the exam each domain covers. Spend your study time in proportion — the heavier the domain, the more questions you'll see.
Key terms
- GitHub Copilot
- GitHub Copilot is an AI pair-programming tool that suggests code and answers questions inside the IDE, CLI, and github.com. GH-300 centers on using it productively, responsibly, and securely.
- Copilot Chat
- Copilot Chat is the conversational interface to GitHub Copilot for asking questions, explaining code, and generating changes. GH-300 covers its commands, limits, and prompt-file reuse.
- Agent Mode
- Agent Mode is a GitHub Copilot capability that autonomously plans and carries out multi-step coding tasks across files. GH-300 covers Agent Mode, Copilot Edits, and delegating to sub-agents.
- Copilot Edits
- Copilot Edits is a GitHub Copilot feature that applies AI-suggested changes across multiple files in a single flow. GH-300 covers it among Copilot's development features.
- MCP
- MCP (Model Context Protocol) is an open standard that lets AI tools like Copilot connect to external tools and data sources. GH-300 covers using MCP to enhance development workflows.
- GitHub Copilot CLI
- GitHub Copilot CLI is a command-line interface that brings Copilot assistance to the terminal for generating scripts and managing files. GH-300 covers installing and using it interactively and in sessions.
- Prompt engineering
- Prompt engineering is the practice of structuring prompts and context to get better output from Copilot. GH-300 dedicates a domain to prompt engineering and context crafting, including zero-shot and few-shot prompting.
- Responsible AI
- Responsible AI is the practice of using AI ethically and safely, accounting for its risks and limitations. GH-300 requires understanding responsible AI principles and validating Copilot output.
- Content exclusion
- Content exclusion is a GitHub Copilot setting that prevents specified files or repositories from being used as context. GH-300 covers configuring content exclusions for privacy and safeguards.
- Public code filter
- The public code filter is a Copilot safeguard that blocks suggestions matching public code, reducing licensing and duplication concerns. GH-300 covers enabling it under privacy and safeguards.
- Inline suggestion
- An inline suggestion is a code completion that GitHub Copilot proposes directly in the editor as you type. GH-300 covers triggering Copilot through inline suggestions, chat, CLI, and agent mode.
- Code suggestion lifecycle
- The code suggestion lifecycle is the flow by which Copilot turns editor context into a filtered, post-processed suggestion. GH-300 covers visualizing this lifecycle in the data and architecture domain.
- Audit log
- An audit log records Copilot-related events for an organization, supporting governance and compliance. GH-300 covers using audit log events when managing organization-wide settings and policies.