GH-300 Practice Questions

The organization audit log for GitHub Copilot Business is designed to track significant administrative and policy-related events, rather than individual user interactions or operational outputs. Changes to content exclusion settings represent a critical policy modification that dictates which files and content GitHub Copilot should ignore. This action directly impacts security, compliance, and data privacy for the organization. Therefore, GitHub explicitly records the copilot.updatecontentexclusionsettings event in the audit log to provide administrators with a clear history of policy changes.

A. Accepted chat suggestions: This is a user-level interaction. The organization audit log tracks administrative policy changes, not the acceptance of individual suggestions by users.

B. Code suggestions made by GitHub Copilot: This is a core operational function. Logging every single suggestion would create an unmanageable volume of data and is not the purpose of an audit log.

D. Suggestions blocked by duplication detection filtering: This is a real-time filtering action. The audit log tracks changes to the policy setting for this filter, not each instance of the filter blocking a suggestion.

1. GitHub Docs. "Audit log events for your organization." GitHub

Inc. Accessed May 2024. In the "Copilot category" section

the documentation explicitly lists copilot.updatecontentexclusionsettings as an event that is recorded in the audit log. This event is triggered "When the content exclusion settings for GitHub Copilot are updated." The other options (A

B

D) are not listed as auditable events.

GitHub Copilot Chat employs a multi-source strategy to generate responses. Its foundation is a large language model (LLM) trained on a massive dataset of public source code and text. To provide relevant and specific answers, it analyzes the context of the user's current project, including the code in open files within the IDE. For questions about recent developments, new libraries, or topics outside its training data, Copilot Chat can leverage the Bing search engine to access and synthesize current information from the web. This combination of pre-trained knowledge, real-time code context, and external search capabilities allows it to provide comprehensive and accurate assistance.

A. This is incomplete. Copilot Chat enhances its base model's knowledge with real-time user code context and external web search results, which are critical features.

B. This is inaccurate. Bing search is a supplementary tool for current information, not the primary source for all responses, which fundamentally rely on the trained model.

D. This is incorrect. While Copilot can use context from a user's workspace, its core knowledge comes from its vast training data, not exclusively from user-provided documents.

---

1. GitHub Docs

"About GitHub Copilot Chat": This official documentation states

"Copilot Chat can... use Bing search to help answer questions about the latest software development trends." This confirms the use of external sources. (Source: GitHub

Inc.

GitHub Docs

"About GitHub Copilot Chat"

Accessed 2024).

2. GitHub Docs

"Using GitHub Copilot Chat in your IDE": This page explains the use of user code as context: "Copilot Chat is aware of the code you have typed and any error messages that are showing in the IDE. This means you can ask specific questions about your code and Copilot Chat will provide answers based on the context of your project." (Source: GitHub

Inc.

GitHub Docs

"Using GitHub Copilot Chat in your IDE"

Section: "Asking questions about your code"

Accessed 2024).

3. GitHub Copilot Trust Center

"Privacy": This resource details the training data: "GitHub Copilot is powered by generative AI models developed by GitHub

OpenAI

and Microsoft... The model is trained on natural language text and source code from publicly available sources

including code in public repositories on GitHub." This confirms the use of a foundational training dataset. (Source: GitHub

Inc.

GitHub Copilot Trust Center

"Privacy"

Accessed 2024).

GitHub Copilot Business provides administrators with the ability to manage which files Copilot can use for code suggestions. This is achieved by configuring "content exclusions" within the repository's settings on the GitHub.com user interface. Administrators can specify paths to files or directories that should be ignored by Copilot. The content of these specified files will not be used to inform code completion suggestions, ensuring that sensitive or irrelevant code is excluded from the model's context.

A. The .gitignore file is used by the Git version control system to determine which files or directories should be ignored and not tracked, it does not control GitHub Copilot's behavior.

B. A copilot.ignore file is not a recognized or supported configuration file for managing GitHub Copilot's content exclusions.

D. Renaming files is not a standard or scalable method for managing content exclusions; it is a manual workaround, not a built-in feature.

---

1. GitHub Docs. (n.d.). Configuring content exclusions for GitHub Copilot. GitHub. Retrieved from https://docs.github.com/en/copilot/managing-copilot-business/configuring-content-exclusions-for-github-copilot.

Reference Details: The section "Configuring content exclusions in your repository settings" explicitly states: "You can configure GitHub Copilot to ignore files in your repository. You do this by specifying excluded paths in the repository's settings... On GitHub.com

navigate to the main page of the repository. Under your repository name

click Settings... In the 'Code & automation' section of the sidebar

click Copilot." This confirms that the user interface settings are the correct method.

GitHub Copilot's effectiveness is highly dependent on the quality and relevance of the context it receives. The primary sources of context are the files currently open in the Integrated Development Environment (IDE). By opening files relevant to your current task, you provide Copilot with a broader understanding of the project's architecture, existing functions, and coding conventions. Additionally, when interacting with Copilot Chat or using inline prompts, providing specific code snippets directly within the prompt gives Copilot precise, immediate context to generate a targeted and accurate response, such as refactoring a function or explaining a block of code.

C. The .gitconfig file is used for configuring the Git version control system and has no influence on the context used by GitHub Copilot.

D. Simply adding file paths to a prompt is not the standard or most effective method; Copilot gains context by analyzing the actual content of open files.

1. Official GitHub Documentation

"Getting started with GitHub Copilot": This document explains the core functionality of Copilot. It states

"As you type

it analyzes the context from your comments and code

and offers suggestions from within your editor... GitHub Copilot processes context from the file you are editing

and related files..." This directly supports that open files (tabs) are a key source of context.

Source: GitHub Docs

"Getting started with GitHub Copilot

" section on "About GitHub Copilot."

2. Official GitHub Documentation

"Using GitHub Copilot Chat in your IDE": This guide details how to interact with Copilot Chat. It emphasizes providing specific context in prompts: "You can ask Copilot Chat to help with a specific piece of code. To do this

highlight the code in your editor and then ask your question." This action is equivalent to adding a relevant code snippet to the prompt's context.

Source: GitHub Docs

"Using GitHub Copilot Chat in your IDE

" section on "Asking questions about your code."

GitHub Copilot custom models are fine-tuned on a selected set of an organization's private repositories. This process allows the model to learn the specific coding conventions, architectural patterns, internal libraries, and idiomatic practices unique to that codebase. As a result, the suggestions provided are more contextually relevant and aligned with the organization's established standards, improving developer productivity and code consistency. The primary benefit is tailoring the AI's output to the specific context of the enterprise's code.

A. Custom models are optimized for relevance and context, not necessarily for speed. The process of routing to a specific fine-tuned model does not inherently make responses faster.

C. The custom model is a fine-tuned version of a base model hosted on GitHub's infrastructure, not a separate Large Language Model (LLM) engine run by the organization.

D. No AI model, including a custom one, can guarantee 100% correctness. Suggestions should always be reviewed by the developer for accuracy, security, and performance.

1. GitHub Docs

"About GitHub Copilot custom models." This document states

"By creating a custom model

you can improve the quality of suggestions for your developers by fine-tuning a model on your own internal source code. The fine-tuned model will have a better understanding of your organization's coding conventions

patterns

and libraries

and will be able to generate more relevant and helpful suggestions." This directly supports option B.

2. GitHub Docs

"Managing Copilot custom models." This page details the process of selecting repositories for fine-tuning

confirming that the model learns from the organization's specific code. It makes no claims about speed (A) or guaranteed correctness (D).

3. GitHub Blog

"GitHub Copilot Enterprise is now generally available." In the announcement

it is highlighted that custom models are "fine-tuned on your company’s codebase

" which enables Copilot to "understand your organization’s institutional knowledge and generate more tailored suggestions." This reinforces that the benefit is about adapting to internal patterns (B)

not using a separate engine (C).

GitHub Copilot Chat assists in ensuring a function's correctness by generating unit tests using commands like /tests. When invoked, Copilot analyzes the selected code's context and semantic meaning to suggest relevant test cases. These generated tests inherently include assertions (e.g., assert in Python, expect in Jest) which are statements that check if a certain condition is true. These assertions are the primary mechanism within the generated test code to verify that the function behaves as expected, thereby helping to ensure its correctness.

B. While Copilot Chat can write tests, it does not guarantee writing all necessary tests for complete coverage. The developer is still responsible for ensuring the test suite is comprehensive.

C. Writing the implementation code is a primary function of Copilot, but this action itself does not ensure correctness. The generated code is what needs to be tested and verified.

D. GitHub Copilot Chat is a code generation tool; it does not have the capability to compile or execute code. Running tests is performed by the developer within their local environment or CI/CD pipeline.

1. GitHub Docs

"Using GitHub Copilot Chat in your IDE": This official documentation describes the /tests slash command

which generates unit tests for selected code. The examples implicitly show that these tests contain assertions to validate functionality. This process relies on understanding the code's context to create meaningful checks. (Reference Section: "Asking GitHub Copilot Chat to write or fix code").

2. GitHub Docs

"Getting started with GitHub Copilot": The documentation explains that Copilot helps developers "write tests" and "debug code

" positioning it as an assistant. The generation of tests with assertions is a core part of this assistance

but it is distinct from executing those tests or guaranteeing complete coverage. (Reference Section: "About GitHub Copilot and IDEs").

GitHub Copilot is trained on a vast corpus of public code, which includes a wide spectrum of quality and security. Consequently, it can inadvertently reproduce patterns from this data that contain known security flaws or do not adhere to modern best practices. Heavy reliance on its suggestions without rigorous review and validation by the developer can lead to the introduction of vulnerable or suboptimal code into a project. The developer remains the ultimate authority and is responsible for the quality, security, and correctness of the final codebase.

B. GitHub Copilot is designed to increase developer velocity by automating repetitive coding tasks and reducing boilerplate, not decrease it.

D. While some suggestions may be irrelevant, the tool's overall purpose and measured impact is to reduce, not increase, development lead time.

1. Official Vendor Documentation: GitHub Copilot Trust Center. In the "Security" and "Code Quality" sections

Microsoft explicitly states

"You are responsible for ensuring the security and quality of your code. We recommend you take the same precautions when using code generated by GitHub Copilot that you would when using any code you didn’t write yourself. These precautions include rigorous testing

IP scanning

and checking for security vulnerabilities." This directly supports the risks outlined in answers A and C.

Source: GitHub. "GitHub Copilot Trust Center." Accessed 2023.

2. Peer-Reviewed Academic Publication: A 2021 study evaluated the security of code generated by GitHub Copilot across 89 different scenarios. The study found that approximately 40% of the generated programs contained security vulnerabilities

often stemming from the top suggestion. This empirically validates the risk of introducing security flaws (Answer A).

Source: Hammond

P.

Pearce

H.

et al. (2021). "Asleep at the Keyboard? Assessing the Security of GitHub Copilot's Code Contributions." 2021 IEEE Symposium on Security and Privacy (SP)

pp. 1-18. DOI: 10.1109/SP40001.2022.00092. (Specifically

see Section IV. FINDINGS

which details the prevalence of vulnerabilities).

3. Peer-Reviewed Academic Publication: Research on the quality of AI-generated code highlights that models like Copilot

trained on public repositories

can suggest code that is outdated

non-idiomatic

or uses deprecated APIs. This directly supports the concern that suggestions may not reflect current best practices or standards (Answer C).

Source: Yetistiren

B.

et al. (2023). "Assessing the Quality of GitHub Copilot's Code Generation." Proceedings of the 20th International Conference on Mining Software Repositories (MSR '23)

pp. 634–645. DOI: 10.1109/MSR59073.2023.00088. (Specifically

see Section 4.2

"Code Quality Analysis

" which discusses issues like outdated practices).

GitHub Copilot for Business and Enterprise offers a specific policy setting to mitigate the risk of IP infringement. This feature, when configured, checks code suggestions against public code available on GitHub. By setting the "Suggestions matching public code" policy to 'blocked', an organization can prevent Copilot from showing suggestions that match existing public code snippets. This is a direct, configurable control provided by Microsoft and GitHub to help organizations manage their intellectual property and compliance requirements when using the AI tool. This filter is a key administrative feature for enterprise governance.

B. Enforce blocking of MIT or GPL licensed code: The filter is not license-specific; it blocks suggestions matching any public code, regardless of its license, to prevent unintentional adoption.

C. You need to check code suggestions yourself before accepting: This is a recommended best practice and a user's responsibility, not a configurable setting that provides automated protection from Microsoft and GitHub.

D. Enable GitHub Copilot license checking: This is not the correct name for the feature. The actual configuration is for filtering suggestions that match public code, not for performing a "license check."

---

1. GitHub Docs

"Configuring GitHub Copilot settings on GitHub.com." Under the section "Managing GitHub Copilot policies for your organization

" the documentation details the "Suggestions matching public code" policy. It states

"You can decide whether to block suggestions from GitHub Copilot that match public code on GitHub... When you choose to block

GitHub Copilot will check code suggestions with their surrounding code of about 150 characters against public code on GitHub. If there is a match or a near match

the suggestion will not be shown." This directly supports option A as the correct configuration.

2. GitHub Docs

"Frequently asked questions about GitHub Copilot." In the "GitHub Copilot and open source" section

it clarifies the purpose of the filter: "We have built a filter to help you detect and suppress suggestions from GitHub Copilot that match public code on GitHub. You can choose to enable or disable the filter." This confirms the existence and purpose of the filter described in option A.

GitHub Copilot Enterprise enhances the pull request review process by leveraging AI to improve comprehension and efficiency. It can automatically generate a summary for a pull request, which includes a high-level prose description and a detailed, bulleted list of the specific code changes. This helps reviewers quickly grasp the purpose and scope of the proposed modifications. Additionally, Copilot Enterprise provides an interactive chat interface directly within the pull request, allowing reviewers and authors to ask specific questions about the changeset to clarify functionality, understand the impact of changes, or get explanations for complex code blocks.

A. It automatically merges pull requests after an automated review.

This is incorrect. Auto-merging is a repository automation feature managed through branch protection rules and merge queues, not a function of GitHub Copilot.

C. It can validate the accuracy of the changes in the pull request.

This is incorrect. Copilot can identify potential issues and suggest improvements, but it cannot validate the logical correctness or business accuracy of code, which requires human judgment.

1. GitHub Docs

"About Copilot pull request summaries." This document states

"Copilot pull request summaries is a feature that allows you to create a summary of the changes made in a pull request... The summary is generated by Copilot based on the commit messages and the code changes in the pull request. It consists of a prose summary and a bulleted list of the changes." This directly supports option B.

2. GitHub Docs

"Using GitHub Copilot for Pull Requests." This official documentation explains

"GitHub Copilot for Pull Requests is a feature that's powered by AI and suggests descriptions for your pull requests... It can also help you understand a pull request by allowing you to ask questions about it." This directly supports option D.

3. GitHub Docs

"Managing a merge queue." This page details GitHub's native auto-merge functionality

which is distinct from Copilot. It states

"A merge queue helps you increase velocity by automating the process of merging pull requests into a busy target branch." This confirms that auto-merging is a separate platform feature

supporting the exclusion of option A.

4. GitHub Docs

"About GitHub Copilot Enterprise." The overview of Copilot Enterprise features highlights its role in "understanding and navigating codebases" and "summarizing pull requests

" but makes no claim about validating the logical accuracy of code

which supports the exclusion of option C.

Zero-shot prompting is a technique where a large language model, like GitHub Copilot, is given a task or question directly without any preceding examples to guide its response. The model is expected to understand the request and generate a relevant output based solely on its vast pre-trained knowledge. This method tests the model's ability to generalize to new, unseen tasks. For instance, simply asking "Write a Python function to calculate the factorial of a number" without showing a sample function is a zero-shot prompt.

A. This is too vague. While a zero-shot prompt has less context than a few-shot prompt, it still requires sufficient context to define the task clearly.

B. This describes an instruction for output quality or constraint, not a specific prompting methodology like zero-shot.

D. This describes few-shot prompting, where one or more examples are provided in the prompt to guide the model's output.

E. This is a more detailed form of few-shot prompting, providing specific implementation examples to constrain the model's approach.

---

1. Microsoft Learn

"Introduction to prompt engineering": In the section "Prompting techniques

" it defines zero-shot prompting: "With this approach

you provide a prompt to the model without any examples... the model is expected to understand and answer the question based on its existing knowledge."

Source: Microsoft Azure AI Documentation

"Introduction to prompt engineering

" Section: "Prompting techniques."

2. Brown

T. B.

et al. (2020). Language Models are Few-Shot Learners. Advances in Neural Information Processing Systems

33

1877-1901. This foundational paper defines the concept in the context of large language models.

Section 2.2

"Few-shot

one-shot

and zero-shot": "In the zero-shot setting... we give the model a natural language description of the task and a prompt... The model is not given any examples."

DOI: https://doi.org/10.48550/arXiv.2005.14165

3. GitHub Docs

"Getting started with GitHub Copilot": The documentation implicitly describes zero-shot prompting when instructing users to write clear

descriptive comments or function signatures to get code suggestions

without mentioning the need for prior examples within the same prompt.

Source: GitHub Docs

"Using GitHub Copilot

" Section: "Getting started with GitHub Copilot." The fundamental usage pattern described is zero-shot.