Hello, you are using an old browser that's unsafe and no longer supported. Please consider updating your browser to a newer version, or downloading a modern browser.
GitHub Copilot is the GitHub certification for using Copilot well and governing it for a team, and this guide covers it as of October 2026. You'll find the GH-300 exam, the six domains and their weights, the format and passing score, what the August 2026 update changed, how the two-year renewal transition works, and the roles it fits.
GitHub Copilot, earned by passing exam GH-300, is the GitHub certification for using Copilot to improve software development productivity, quality and security. It tests how you work with Copilot in the IDE, in the CLI and on github.com, how Copilot handles your code and prompts, and how an organization controls it through policies and content exclusions.
The credential is GitHub-branded, but Microsoft administers it. You register with a Microsoft Learn account, schedule through Pearson VUE at a test center or online, and the certification record lives in your Microsoft Learn profile. Microsoft rates it Intermediate, and the title is GitHub Copilot rather than a Microsoft Certified credential.
GH-300 has no required prerequisites. Microsoft expects candidates to be familiar with GitHub fundamentals and to have experience with one or more programming languages. The skills measured outline is current as of August 7, 2026 and is published on Microsoft Learn.
Generative AI risks and limits, ethical use, harms and mitigation, validating output. 15 to 20%.
IDE, chat, CLI, agent mode, MCP, code review, Spaces, Spark, org policies and audit logs. 25 to 30%.
Data flow, prompt building, proxy filtering, post-processing, suggestion lifecycle, LLM limits. 10 to 15%.
Prompt structure, how context is determined, zero-shot and few-shot prompting, chat history. 10 to 15%.
Code generation, refactoring, documentation, tests, legacy modernization, security suggestions. 10 to 15%.
Content exclusions, editor settings, output ownership, public code filter, troubleshooting. 10 to 15%.
Four reasons GH-300 earns a place on the plan for anyone who uses or manages Copilot in 2026.
GitHub's certification list on Microsoft Learn now runs to seven, counting a business-focused agentic AI exam that is still in beta. GH-300 is the one that tests Copilot directly: in the IDE, in Copilot CLI, on github.com, and as a service an organization licenses and controls. GH-900 touches Copilot plans lightly, and GH-600 assumes you already know what GH-300 tests.
Microsoft tags GH-300 for app makers, developers, DevOps engineers and technology managers. The audience profile asks for GitHub fundamentals and at least one programming language, and the policy, audit log and subscription objectives are written for the team lead who answers for Copilot use.
The August 7, 2026 outline names Copilot CLI, agent mode, Copilot Edits, MCP, agent sessions and sub-agents, Spaces, Spark and code review policies. Our guide to the 2026 GitHub certification track shows how GH-300 fits with the other exams.
GH-300 is valid for two years. GitHub is moving its certifications onto Microsoft's recertification process so holders can renew without retaking the full exam, and certifications that expire before that process is ready are being extended by six months. Neither GitHub nor Microsoft claims a DoD 8140 mapping for it. The current matrix is published at the DoD Cyber Exchange.
Its weight is practical instead. Copilot sits inside the editor, the pull request and the command line, so a team that licenses it needs people who can use it well, check what it writes and set its policies. GH-300 is the credential that shows you can do all three.
The certification, the GH-300 exam, and what happens after you pass, as of October 2026.
GH-300 sits in the middle of the GitHub track. The right next exam depends on what you answer for at work, and these three build most directly on what GH-300 covers.
The next step once agents act on their own in your workflow. GH-600 leads to GitHub Certified: Agentic AI Developer and covers six skill areas, from agent architecture and tool use to memory, multi-agent orchestration and guardrails. It runs 120 minutes, and Microsoft's audience profile expects experience with coding agents including GitHub Copilot.
For whoever owns Copilot settings across an enterprise. The GH-100 study guide now carries the title GitHub Enterprise Administrator, and its July 2026 outline covers identities and access, the enterprise environment, secure development and compliance, Actions, and usage monitoring. The policy and audit log objectives in GH-300 are a preview.
For developers and AppSec engineers who want the security side. The July 2026 GH-500 outline covers Secret Protection, supply chain security and Code Security with CodeQL, and the Code Security domain includes autofix capabilities for code scanning alerts. The GH-300 objective on suggesting security improvements is where that work starts.
GH-300 sits at the intermediate tier of the GitHub track. GH-900 is optional groundwork below it, and from here the track branches into AI agents with GH-600 or into enterprise administration, automation and security.
Optional groundwork
Use it and govern it
AI agents, enterprise or security
Two questions to answer before you commit: can you sit it, and should you pursue GH-300 specifically. Here's a straight answer to both.
No required prerequisites.
GH-300 has no prerequisite certification and no enforced experience requirement. You create a Microsoft Learn account, register with a personal Microsoft account rather than a work account so the record stays with you when you change jobs, and book through Pearson VUE up to 90 days ahead. Microsoft does expect GitHub fundamentals and experience with at least one programming language.
Part of the exam lives in Copilot Business and Enterprise settings.
A personal Copilot plan gives you suggestions, chat and the CLI to practice with. It won't show you organization-wide policy management, Copilot Code Review policies, audit log events or seat management through the REST API, because GitHub puts centralized management and policy control in Copilot Business and Copilot Enterprise. Get time in an organization that licenses Copilot, or train somewhere that provides one, before you book.
GH-300 supports a role rather than qualifying you for one on its own. These are the six roles where using and governing Copilot is part of the job description.
Writes code with Copilot every day. GH-300 adds the parts the editor never shows you, such as how a prompt is built from context, what proxy filtering does, and where large language models get suggestions wrong.
Writes scripts and manages files from the terminal, which is the work the Copilot CLI objectives describe. GH-300 also covers agent mode and sub-agents, and GH-200 GitHub Actions is the natural next exam for this role.
Decides how a team uses Copilot and what gets reviewed before merge. The organization policy, Copilot Code Review policy and audit log objectives in GH-300 map to those decisions, and Microsoft tags the exam for technology managers.
Manages Copilot seats, policies and content exclusions for an organization or enterprise. GH-300 covers managing subscriptions through the REST API, and GH-100 GitHub Administration takes the enterprise side further.
Uses Copilot to draft unit and integration tests, find edge cases and write assertions. The productivity domain covers that work directly, and the responsible use domain explains why every generated test still needs a human check.
Runs a company's Copilot rollout, writes the instructions files and prompt files teams share, and trains people on prompt crafting and responsible use. GH-300 covers each of those topics, and GH-600 is the step once agents act inside the pipeline.
GH-300 sits between GitHub's fundamentals exam and its agentic AI exam, and it often gets weighed against AI credentials from the cloud vendors. Here's how it lines up against GH-900, GH-600 and the AWS Certified AI Practitioner.
| GitHub Copilot (GH-300) | GitHub Foundations (GH-900) | Agentic AI Developer (GH-600) | AWS AI Practitioner (AIF-C01) | |
|---|---|---|---|---|
| Issuer | GitHub, administered by Microsoft | GitHub, administered by Microsoft | GitHub, administered by Microsoft | AWS |
| Level | Intermediate | Fundamentals | Intermediate | Foundational |
| Focus | Using and governing Copilot | Working in GitHub | Running AI agents in the SDLC | AI, ML, generative AI and AWS AI services |
| Current Outline | August 7, 2026 | January 2026 | See Microsoft Learn | Exam guide v1.4 |
| Exam Time | 100 min | 100 min | 120 min | 90 min |
| Passing Score | 700 of 1000 | 700 of 1000 | 700 of 1000 | 700 of 1000 |
| Cost (US) | $99 | $99 | $99 | $100 |
| Validity | 2 years | Does not expire | 2 years | 3 years |
| Best For | Developers and team leads using Copilot | New GitHub users and non-developer roles | Engineers running agents in real workflows | People who use AI on AWS in cloud, data and business roles |
Microsoft prices GitHub exams by the country where you test, so fees vary outside the United States. GH-600 is offered in English only. None of the four is mapped to a DoD 8140 work role by its vendor, so their recognition is market-driven. AWS figures reflect the Training Camp AWS AI Practitioner guide, and the two-year GitHub credentials are moving onto Microsoft's recertification process, so check your expiration date in your Microsoft Learn profile.
Certification Matchups: the published facts, strengths and weak spots of 33 certifications, side by side.
As a Microsoft training partner, we teach GH-300 in our GitHub AI Certification Track Boot Camp, a three-day live course, online or in person, that pairs it with GH-600 from Official Microsoft Courseware. You work in a hosted GitHub organization with Copilot and agent features turned on, sit full-length practice exams, and get vouchers for both exams plus a retake. Teams that run GitHub Enterprise can take GH-300 in our five-day GitHub Enterprise Certification Track instead, alongside GH-100, GH-200 and GH-500.
The full GitHub track, the risks that come with agents, other vendors' AI credentials, and how to study from an exam outline.
The structure of each GitHub exam as of September 2026, what the August 2026 GH-300 update touched, why the data and architecture domain catches daily Copilot users out, and the order that makes sense by role.
The ten risks OWASP lists for AI systems that plan, remember, call tools and act with delegated authority. Worth reading before you switch on Copilot agent mode for a whole team.
What the four Claude certification exams cover, how they are delivered and scored, and why only people at Claude Partner Network organizations can sit them today. Another vendor's approach to certifying AI-assisted work.
The AI governance, security, audit and vendor certifications on offer in 2026, who each one is written for, and the Microsoft AI exams that were renumbered this year.
How Microsoft certification moved from product exams to job roles, why 2026 brought so many retirements, and how GitHub's exams came onto Microsoft Learn and Pearson VUE in 2025.
The Microsoft certifications worth knowing this year, why so many popular-cert lists are out of date, and where the new Copilot and agent exams fit on the Microsoft 365 and Azure side.
How to read an exam outline, what the verbs in each objective mean, and how to turn domain weights into a study plan. The same method works on the GH-300 study guide.
The GH-300 exam is organized into six domains, each carrying its own weight. Use GitHub Copilot features is the heaviest at 25 to 30 percent, responsible use comes next at 15 to 20 percent, and the other four sit at 10 to 15 percent each. Click any domain for what it covers.
The risks and limitations of generative AI tools, ethical and responsible use, and the harms AI use can cause along with ways to reduce them. Also covers why AI output has to be validated and how to operate Copilot responsibly in daily work. It outweighs four of the other five domains, so give it more study time than its abstract topics suggest.
Enabling Copilot in the IDE and triggering it through inline suggestions, chat, the CLI and agent mode, plus installing GitHub Copilot CLI and using it interactively and in sessions. Also covers Copilot Edits, MCP, agent sessions and delegating work to sub-agents, Copilot code review, Spaces, Spark, pull request summaries, instructions files and prompt files. The organization side sits here too: policy management, Copilot Code Review policies, audit log events and managing subscriptions through the REST API. The heaviest domain on the exam.
How Copilot uses, moves and shares data, how it processes input and builds a prompt, and what proxy filtering and post-processing do to a response. Also covers the lifecycle of a code suggestion and the limitations of large language models and of Copilot itself. Daily use doesn't teach this material, so plan to read GitHub's documentation on how Copilot handles data.
How a prompt is structured, what context Copilot draws on and how that context is determined, and when zero-shot or few-shot prompting fits. Also covers prompt crafting best practices, prompt engineering principles, and how the prompt process flow uses chat history.
Using Copilot for code generation, refactoring and documentation, to learn faster with less context switching, to generate sample data and to modernize legacy code. Also covers generating unit and integration tests, identifying edge cases and writing assertions, and having Copilot suggest security improvements and performance optimizations.
Configuring content exclusions and editor settings, and the ownership and limitations of what Copilot produces. Also covers enabling the filter for suggestions that match public code, and resolving problems with suggestions and content exclusions. Content exclusions also appear under the features domain, so expect the topic from two angles.
Domains and weights reflect the Microsoft Learn GH-300 study guide, skills measured as of August 7, 2026. Microsoft says most questions cover generally available features, but the exam may include Preview features that are commonly used, so confirm the current outline on Microsoft Learn before scheduling.
The questions candidates ask most often when researching the GitHub Copilot certification.
GitHub Copilot, earned by passing exam GH-300, is the GitHub certification for using Copilot to improve software development productivity, quality and security. It covers responsible AI use, Copilot features across plans, data and architecture, prompt engineering, developer productivity, and privacy safeguards. Microsoft rates it Intermediate and delivers the exam on GitHub's behalf.
Microsoft administers it, but the credential stays GitHub-branded. You register with a Microsoft Learn account, schedule through Pearson VUE, and the certification appears in your Microsoft Learn profile, but the title is GitHub Copilot rather than a Microsoft Certified credential. Training Camp teaches it as a Microsoft training partner using Official Microsoft Courseware.
No. GH-300 has no required prerequisite and no enforced experience requirement, so anyone can register. Microsoft does say candidates should be familiar with GitHub fundamentals and have experience with one or more programming languages. If GitHub itself is new to you, GH-900 GitHub Foundations is the gentler place to start.
The six domains are Use GitHub Copilot responsibly (15 to 20 percent), Use GitHub Copilot features (25 to 30 percent), Understand GitHub Copilot data and architecture (10 to 15 percent), Apply prompt engineering and context crafting (10 to 15 percent), Improve developer productivity with GitHub Copilot (10 to 15 percent), and Configure privacy, content exclusions, and safeguards (10 to 15 percent). Features carries the most weight by a clear margin.
Microsoft does not publish a question count for GH-300. You have 100 minutes, and Microsoft says the exam may include interactive components. Unlike Microsoft's associate and expert exams, GitHub exams don't give you access to Microsoft Learn during the test, so plan to answer everything from what you already know.
You need 700 on a scale of 1 to 1000. Microsoft uses scaled scoring, so 700 doesn't mean 70 percent correct, and the passing line accounts for the difficulty of the questions you were given. If you fail a first attempt, you can retake after 24 hours, with longer waits for later attempts.
The GH-300 exam costs $99 USD in the United States as of October 2026, scheduled through Pearson VUE. Microsoft prices by the country where you test, so check the exam page for your region. Training Camp's GitHub AI Certification Track Boot Camp includes vouchers for both GH-300 and GH-600 in the program fee.
Microsoft's change log shows a minor update on August 7, 2026. Three objective groups changed (Use GitHub Copilot in the IDE, Use GitHub Copilot features and capabilities, and Apply safeguards and troubleshoot), while the audience profile and their parent domains are marked unchanged. The current outline names Copilot CLI, agent mode, MCP, agent sessions and sub-agents, Spaces and Spark, so confirm the skills measured date on any study material before you rely on it.
No DoD 8140 mapping is claimed for it. Neither GitHub nor Microsoft lists GH-300 against a DoD 8140 work role, so don't count on it to satisfy a cyber-coded position requirement. Its value is practical instead, since it shows you can use Copilot well and govern it for a team. See the full DoD 8140 work role paths.
Yes. GitHub certifications are valid for two years. GitHub is moving them onto Microsoft's recertification process, which will let holders renew without retaking the full exam, and certifications that expire before that process is available are being extended by six months. If yours has already lapsed, email learn@github.com and GitHub will provide one exam voucher for your first renewal attempt.
GH-300 fits software developers, DevOps engineers, engineering managers and team leads, GitHub platform administrators who manage Copilot seats and policies, test automation engineers, and the people who run a company's Copilot rollout. Microsoft tags it for app makers, developers, DevOps engineers and technology managers. It supports a role rather than qualifying you for one on its own.
Pick by what you answer for. If agents now act inside your pipelines, GH-600 Agentic AI Developer is the next step, and its audience profile assumes Copilot experience. Enterprise administrators take GH-100, application security engineers take GH-500, and pipeline owners take GH-200. Product and program managers can look at GH-700, which leads to GitHub Certified: Agentic AI Business Professional and is in beta as of October 2026. Training Camp's GitHub AI Certification Track pairs GH-300 with GH-600, and the GitHub Enterprise Certification Track covers GH-100, GH-200, GH-300 and GH-500.
For developers and team leads who use Copilot every day, yes. It costs $99, the August 2026 outline tracks current features such as Copilot CLI and agent mode, and its data, policy and content exclusion objectives cover what the editor never shows you. If GitHub itself is new to you, start with GH-900. If you already run agents in production workflows, GH-600 may match your job more closely.
If you're weighing the certification, planning GH-300 and GH-600 for yourself, or arranging Copilot training for a team, tell us where you are and we'll help you map out the right path.