Apply modern development practices

The exam tests the purpose and boundaries of GitHub's automation, AI, and cloud-development tools, not a specific programming language.

Official GitHub Docs visual: the Codespaces architecture.

Tool	Primary purpose	Key distinction
GitHub Actions	Automate workflows such as build, test, and deployment	Workflow automation triggered by repository events
GitHub Copilot	AI assistance for coding and related tasks	Suggestions and chat assistance, with plan and policy differences
GitHub Codespaces	Cloud-hosted development environment	Full configured environment, often using dev containers
github.dev	Browser editor for quick edits and exploration	Lighter-weight editor, not a hosted compute environment like Codespaces
Dev container	Configuration for a repeatable development environment	Makes tools and setup consistent across developers

flowchart TD A[Repository event] --> B[GitHub Actions workflow] B --> C[Build or test] C --> D[Report result] E[Developer] --> F[Codespaces or local dev container] F --> G[Edit and test] G --> H[Pull request] I[Copilot] -. assists with .-> G

Original visual synthesizing the modern-development tools listed in the GH-900 study guide.

Compare Copilot offerings

Offering	Study prompt
Individual	What capabilities and billing model are intended for an individual user?
Business	What organization controls and policy needs are introduced for teams?
Enterprise	What additional enterprise governance and knowledge-context options matter?

Use the official documentation for Actions, Copilot, and Codespaces. Product availability can change, so prioritize current documentation over memorized product details.

GitHub Actions fundamentals

GitHub Actions automates repository-centered work. A workflow is defined in the repository and runs when a configured event or schedule occurs.

Term	Meaning	Study cue
Workflow	An automated process defined in a repository	It can build, test, scan, release, or deploy.
Event	Something that can trigger a workflow	Examples include push, pull request, schedule, or manual dispatch.
Job	A group of steps that run together	A workflow can contain one or more jobs.
Step	An individual task in a job	A step can run a command or use an action.
Action	A reusable unit of automation	Actions can be used as part of a workflow step.
Runner	The machine that executes a job	The runner provides the execution environment.

Automation scenario guide

Need	Likely GitHub capability
Run tests every time a pull request changes	An Actions workflow triggered by pull-request activity.
Publish a release after a tagged commit	An Actions workflow with a relevant release or tag trigger.
Perform a repeated repository task on a schedule	A scheduled Actions workflow.
Give a developer a consistent cloud environment	Codespaces with dev-container configuration.
Get AI-assisted suggestions while coding	GitHub Copilot in a supported environment.

GitHub Copilot concepts

Copilot is an AI-powered assistant that can help users generate, explain, transform, and reason about code and related tasks. It is assistance, not an automatic replacement for review, testing, security checks, or ownership.

Capability named by the objective	What to understand
Code suggestions	Copilot can propose code while a user works in a supported editor.
Chat	A user can ask questions, request explanations, or iterate on a task conversationally.
Agents and Agent Mode	Agent capabilities can help work through a task using an iterative tool-assisted workflow; users remain responsible for review.
Multi-model support	Available model choices can affect how users interact with supported Copilot experiences.
Policy management	Organizations can govern use through plan and organization settings.

Copilot plan comparison

Plan family	Primary study distinction
Individuals	Intended for individual use and personal account context.
Business	Adds organization-oriented management for team use.
Enterprise	Adds enterprise-scale governance and capabilities for organizations with broader needs.

Do not memorize commercial details that can change. Understand why team and enterprise contexts introduce centralized policy and governance needs.

Codespaces, dev containers, and github.dev

Tool	Best use	Compute environment	Configuration concept
Codespaces	Develop, run, and test in a hosted environment	Cloud-hosted environment	Can use a dev container to define tools and setup.
Dev container	Standardize environment setup	Used by supporting development environments	Configuration describes the environment and dependencies.
github.dev	Browse and make quick edits in a browser	Lightweight editor experience	Not a substitute for a full Codespaces compute environment.

Environment workflow

1. Open a repository in Codespaces when the work needs a configured environment.
2. Use or update dev-container configuration when the team needs repeatable tools and dependencies.
3. Make, run, and test changes in the environment.
4. Commit and push changes to a branch.
5. Open a pull request for review and validation.

Readiness check

• Explain workflow, event, job, step, action, and runner in one sentence each.
• Choose between Actions, Copilot, Codespaces, a dev container, and github.dev for a scenario.
• Explain why an AI suggestion still needs human review and repository checks.
• Describe the governance difference between an individual and an organization-managed Copilot context.

Suggested answers

1. Actions vocabulary: A workflow is the automated process, an event starts it, a job is a related set of work, a step is one task in that job, an action is reusable automation, and a runner executes the job.
2. Tool selection: Use Actions for event-driven automation, Copilot for AI-assisted coding work, Codespaces for a configured cloud development environment, a dev container to describe repeatable environment setup, and github.dev for quick browser editing or exploration.
3. Human responsibility: AI suggestions can be incomplete, incorrect, insecure, or unsuitable for a repository. People must review the change, run appropriate tests, and follow the repository's pull-request and security requirements.
4. Governance: An individual uses Copilot in a personal context. An organization-managed context adds centralized policy, administration, and controls that help a team govern use consistently.