GH-300 domain - 28% of the exam

Use GitHub Copilot features

Use GitHub Copilot features is 28% of the GH-300 exam. These are the objectives it covers, each with practice questions, with every answer explained.

The domain in numbers

77
Practice questions
6
Objectives
13
Easy
52
Medium
12
Hard

Where people go wrong

  • Mixing up Chat, Edit Mode, Agent Mode and Plan Mode

    The wrong answers pick the surface that sounds closest to the task. Agent Mode is offered where the developer wants to review each change, Edit Mode where tests must run on their own, and Plan Mode where code must be written. Chat explains without changing files, Edit Mode applies changes for edit-by-edit review, Agent Mode edits files, runs tests and iterates autonomously, and Plan Mode only produces an ordered breakdown before implementation.

    See it in a sample question
  • Placing a feature on the wrong Copilot tier

    The wrong answers name a tier that is too low or too high. Copilot Free and Copilot Pro are individual plans with no organisation seat pool, audit log or central policy. Copilot Business is where REST API seat management, the audit log and the public-code matching policy first appear, so naming Enterprise overspends. Copilot Enterprise adds knowledge bases grounded in organisation documentation, which Business does not offer.

    See it in a sample question
  • Confusing instructions, Spaces, MCP and content exclusions

    These four features are offered in place of one another. Custom instructions are static standing guidance applied to prompts automatically. A Copilot Space is a curated, shared set of repositories and documents that grounds Chat. A Model Context Protocol server gives Chat live access to an external tool and its data. Content exclusions stop named files being used as context, and they never add access to anything.

    See it in a sample question
  • Inventing a separate licence for the Copilot CLI

    The wrong answers attach an extra requirement to the GitHub Copilot CLI: an Enterprise plan, a business tier, a distinct paid terminal licence, or the free tier only. The CLI shares the existing Copilot entitlement used for editor completions and chat. After installation, the remaining step is to sign in with the GitHub account that holds an active Copilot subscription. Content exclusions and a Model Context Protocol server are not prerequisites.

    See it in a sample question

Objectives in this domain

What to study

The official documents this domain's practice questions cite most.

Sample questions from this domain

Free sampleUse GitHub Copilot featuresmedium

A developer has used Copilot Plan Mode to produce an agreed, ordered breakdown of the steps and files for a new feature. They now want Copilot to carry out that plan end to end, including making the edits and running the tests, with minimal step-by-step intervention. What is the appropriate next move?

  • AStay in "Plan Mode" and instruct Copilot to begin applying each planned step directly, since the mode that produced the plan is best placed to execute it.
  • BSwitch to "Edit Mode" and approve each individual edit one by one through the whole feature, which is the only way to apply a multi-step plan.
  • CHand the agreed plan to "Agent Mode" so Copilot executes the steps autonomously, editing the files and running the tests with minimal intervention. Correct
  • DPaste the plan into inline "ghost-text" completions in each file so Copilot fills in the implementation as the developer types through every step.
Recognise that a Plan Mode breakdown is executed by handing it to Agent Mode, which acts autonomously across files and tests. Plan Mode and Agent Mode are complementary: Plan Mode produces an agreed sequence of steps before any code is written, and Agent Mode then performs those steps autonomously, including editing files and running tests, which is exactly the workflow the developer describes.

Why A is wrong: Plan Mode is for analysing and breaking down work before coding, not for autonomous execution, so it is tempting but does not carry out the edits and test runs.

Why B is wrong: Edit Mode requires per-edit review and does not run tests autonomously, which conflicts with the stated wish for minimal step-by-step intervention.

Why C is correct: Agent Mode is the surface built to execute multi-step work autonomously, so passing the Plan Mode breakdown to it matches the goal of carrying out the plan end to end.

Why D is wrong: Inline completions only suggest code at the cursor as a developer types and cannot orchestrate a multi-file plan or run tests, so they cannot execute the plan.

Free sampleUse GitHub Copilot featuresmedium

A team lead is evaluating GitHub Copilot tiers and wants the plan that adds knowledge bases built from organisation documentation plus Copilot answers grounded in those internal repositories. Which plan should the team lead choose?

  • ACopilot Business, because building knowledge bases from internal repositories is one of the seat-management features unlocked at the organisation entry tier.
  • BCopilot Pro, because the paid individual plan grounds Copilot Chat in any private repositories the developer is permitted to read.
  • CCopilot Enterprise, because it layers organisation knowledge bases and repository-grounded chat answers on top of the controls already present in Copilot Business. Correct
  • DCopilot Free, because the no-charge tier indexes organisation documentation to ground its limited monthly chat responses.
Recognise that organisation knowledge bases and repository-grounded chat answers are a Copilot Enterprise capability. Knowledge bases assembled from selected organisation documentation, and Copilot Chat answers grounded in the organisation's own repositories, are distinguishing features of Copilot Enterprise and are not present on Free, Pro or Business.

Why A is wrong: Copilot Business adds seat management and policy controls but not knowledge bases; grounding chat in selected organisation documentation is reserved for the Enterprise tier above it.

Why B is wrong: Copilot Pro grounds chat only in the file or repository currently open, not in curated organisation knowledge bases, so it cannot deliver the documentation-wide grounding requested.

Why C is correct: Copilot Enterprise is the tier that adds knowledge bases compiled from selected documentation and chat answers grounded in the organisation's own repositories, exactly the capability described in the stem.

Why D is wrong: Copilot Free offers only capped chat with no organisation indexing or knowledge bases, so it cannot ground answers in internal documentation across repositories.

Free sampleUse GitHub Copilot featuresmedium

A new joiner needs to ramp up on a service and wants one shared place that bundles the relevant repositories, design notes, and a handful of key docs so Copilot Chat grounds its answers on that curated set. Which Copilot feature is intended for this?

  • AWrite a custom instructions file that lists the documents so Copilot reads each linked source live whenever the new joiner asks a question.
  • BStand up a Model Context Protocol server that streams the documents so Copilot fetches them as a tool during each onboarding conversation.
  • CCreate a Copilot Space that gathers the relevant repositories, design notes, and key docs so Chat grounds its answers on that curated collection. Correct
  • DEnable pull request summaries on the repositories so the new joiner reads generated overviews of past changes to learn the service over time.
Select a Copilot Space to bundle curated repositories and documents as shared grounding context for Copilot Chat. Copilot Spaces let a team assemble repositories, notes, and documents into one shared context that Chat uses to ground its answers, which fits curated onboarding, unlike static instructions, live MCP tools, or per-diff summaries.

Why A is wrong: Custom instructions add standing guidance text and do not assemble a browsable, shared collection of repositories and docs for grounding answers.

Why B is wrong: An MCP server connects external tools and is unnecessary here, since the need is a curated reference collection rather than a live tool integration.

Why C is correct: A Copilot Space bundles selected repositories and documents into one shared context that Chat draws on, which matches grounding answers on a curated onboarding set.

Why D is wrong: Pull request summaries describe individual diffs and do not create a single curated space of repositories and docs to ground Chat for onboarding.

Free sampleUse GitHub Copilot featuresmedium

A new team member wants to start using the GitHub Copilot CLI on their workstation. After installing the CLI tooling, what additional step is required before they can receive command suggestions from it?

  • AEnable "content exclusions" for the repository so that the GitHub Copilot CLI is permitted to read the working directory and return suggestions.
  • BConfigure a "Model Context Protocol" server endpoint, because the CLI cannot produce any shell command suggestions until an external context server responds.
  • CAuthenticate the CLI with their GitHub account that holds an active Copilot subscription, so the tool is authorised to request suggestions. Correct
  • DSwitch the account to "Copilot Enterprise" first, since the GitHub Copilot CLI refuses to return suggestions on lower Copilot subscription tiers.
Know that using the GitHub Copilot CLI requires authenticating with a GitHub account holding an active Copilot subscription. Beyond installing the tooling, the GitHub Copilot CLI must be authenticated against a GitHub account that has an active Copilot subscription before it is authorised to request suggestions, which is separate from optional features such as content exclusions or Model Context Protocol servers.

Why A is wrong: Content exclusions restrict which files are used as context rather than enabling the tool, so they are not the step that lets the CLI start returning suggestions.

Why B is wrong: Model Context Protocol extends context with external tools and is optional, so the CLI generates command suggestions without any MCP server configured.

Why C is correct: The GitHub Copilot CLI requires authentication against a GitHub account with an active Copilot subscription before it can request and return command suggestions.

Why D is wrong: The CLI works with an active Copilot subscription and does not demand Enterprise specifically, so requiring that tier misstates the access requirement.

Other domains in this exam

See also the GH-300 cert hub, the study guide, and the cheat sheet.

Examworthy is not affiliated with or endorsed by GitHub. Original, blueprint-aligned practice material only.