Question 190
The current design is: Use a configured custom agent with relevant instructions and tool restrictions for the review task. The prior failure was: A general-purpose agent makes build changes instead of focusing on the requested analysis. Which ONE ongoing control best protects against recurrence of that specific failure?
- Choose the least-privileged Copilot interaction mode for the requested work.
- Keep custom agent scopes explicit and check output against human review requirements.
- Keep branch protections and required checks for agent-created pull requests.
- Review MCP server provenance and per-tool approvals before enabling external actions.
Explanation
Correct Answer
B:Specialized objectives and tool boundaries were not provided. Keep custom agent scopes explicit and check output against human review requirements.
Incorrect Answers
A:Generated code requires independently defined acceptance tests and review against the real program contract. Preventing this failure involves keep custom agent scopes explicit and check output against human review requirements.
C:Keep branch protections and required checks for agent-created pull requests. Preventing this failure involves keep custom agent scopes explicit and check output against human review requirements.
D:Approved dependency sources and lockfiles protect builds against invented or impersonated packages. Preventing this failure involves keep custom agent scopes explicit and check output against human review requirements.
Question 191
A team reports: A product team repeatedly asks Copilot about the same architecture decision record, API contract and issue links across separate conversations. They observe: The assistant guesses an outdated system interface instead of using the current ADR. Which ONE technical decision most directly corrects the cause?
- Select the relevant function in the IDE and ask Copilot Chat for an explanation using that bounded context.
- Create a reusable `.github/prompts/*.prompt.md` template rather than copy prose repeatedly into each chat.
- Curate those approved resources in a Copilot Space when available and ask questions in that bounded shared context.
- Reference the approved file with the CLI's supported `@path` context syntax.
Explanation
Correct Answer
C:Relevant organizational knowledge was not attached to the conversation context.
Incorrect Answers
A:The requested symbol context was ambiguous or not attached to the prompt. Relevant organizational knowledge was not attached to the conversation context.
B:Task-specific prompt guidance was not stored as a reusable artifact. Relevant organizational knowledge was not attached to the conversation context.
D:The actual file context was omitted from the terminal prompt. Relevant organizational knowledge was not attached to the conversation context.
Question 192
A team has implemented the following change: Curate those approved resources in a Copilot Space when available and ask questions in that bounded shared context. The original requirement was: A product team repeatedly asks Copilot about the same architecture decision record, API contract and issue links across separate conversations. Which ONE validation test best demonstrates the intended behavior?
- Ask a factual question about a specific section and verify the file reference was used.
- Ask a question whose answer appears in the attached ADR and compare the quoted section.
- Compare the explanation to the actual control flow and a failing branch test.
- Invoke the prompt file with a test module and confirm the expected headings and input variables are filled.
Explanation
Correct Answer
B:Ask a question whose answer appears in the attached ADR and compare the quoted section.
Incorrect Answers
A:Ask a factual question about a specific section and verify the file reference was used. That check cannot demonstrate ask a question whose answer appears in the attached ADR and compare the quoted section.
C:Generated code requires independently defined acceptance tests and review against the real program contract. That check cannot demonstrate ask a question whose answer appears in the attached ADR and compare the quoted section.
D:Generated code requires independently defined acceptance tests and review against the real program contract. Verification of this result involves ask a question whose answer appears in the attached ADR and compare the quoted section.
Question 193
During troubleshooting, investigators observe: The assistant guesses an outdated system interface instead of using the current ADR. The system context is: A product team repeatedly asks Copilot about the same architecture decision record, API contract and issue links across separate conversations. Which ONE cause most directly explains the observed failure?
- Relevant organizational knowledge was not attached to the conversation context.
- Task-specific prompt guidance was not stored as a reusable artifact.
- The requested symbol context was ambiguous or not attached to the prompt.
- The actual file context was omitted from the terminal prompt.
Explanation
Correct Answer
A:The remedy should therefore be: Curate those approved resources in a Copilot Space when available and ask questions in that bounded shared context.
Incorrect Answers
B:Task-specific prompt guidance was not stored as a reusable artifact. Relevant organizational knowledge was not attached to the conversation context.
C:The requested symbol context was ambiguous or not attached to the prompt. Relevant organizational knowledge was not attached to the conversation context.
D:The actual file context was omitted from the terminal prompt. Relevant organizational knowledge was not attached to the conversation context.
Question 194
The current design is: Curate those approved resources in a Copilot Space when available and ask questions in that bounded shared context. The prior failure was: The assistant guesses an outdated system interface instead of using the current ADR. Which ONE ongoing control best protects against recurrence of that specific failure?
- Version and test prompt files for recurring, bounded engineering workflows.
- Keep Spaces sources maintained and check permissions before sharing sensitive documents.
- Attach explicit approved paths and avoid unnecessary secrets in CLI prompts.
- Provide selected code or explicit file references for codebase explanations.
Explanation
Correct Answer
B:Relevant organizational knowledge was not attached to the conversation context. Keep Spaces sources maintained and check permissions before sharing sensitive documents.
Incorrect Answers
A:Generated code requires independently defined acceptance tests and review against the real program contract. It would leave keep Spaces sources maintained and check permissions before sharing sensitive documents without the specified safeguard.
C:Cryptographic algorithm decisions must follow authoritative security standards rather than unverified model claims. Preventing this failure involves keep Spaces sources maintained and check permissions before sharing sensitive documents.
D:Provide selected code or explicit file references for codebase explanations. It would leave keep Spaces sources maintained and check permissions before sharing sensitive documents without the specified safeguard.
Question 195
A team reports: A new Copilot subscriber follows a dated training exercise that asks them to create a GitHub Spark app in October 2026, but the create-app workflow is unavailable. They observe: The new subscriber cannot create the tutorial’s Spark application even with a valid Copilot subscription. Which ONE technical decision most directly corrects the cause?
- Install the supported GitHub Copilot extension and sign in with the licensed GitHub account in the editor.
- Follow the current GitHub Spark lifecycle notice: do not plan new Spark apps, and choose a supported alternative prototyping workflow.
- Use GitHub Copilot inline code completions from nearby source comments and surrounding code.
- Review local editor settings and the effective Copilot feature enablement for the target language.
Explanation
Correct Answer
B:GitHub Spark stopped accepting new users and new application creation on August 4, 2026.
Incorrect Answers
A:Installing the supported GitHub Copilot extension and sign in with the licensed GitHub account in the editor would not address the need for follow the current GitHub Spark lifecycle notice: do not plan new Spark apps.
C:Using GitHub Copilot inline code completions from nearby source comments and surrounding code would not address the need for follow the current GitHub Spark lifecycle notice: do not plan new Spark apps.
D:Reviewing local editor settings and the effective Copilot feature enablement for the target language would not address the need for follow the current GitHub Spark lifecycle notice: do not plan new Spark apps.
Question 196
A team has implemented the following change: Follow the current GitHub Spark lifecycle notice: do not plan new Spark apps, and choose a supported alternative prototyping workflow. The original requirement was: A new Copilot subscriber follows a dated training exercise that asks them to create a GitHub Spark app in October 2026, but the create-app workflow is unavailable. Which ONE validation test best demonstrates the intended behavior?
- Begin a descriptive function and inspect the ghost-text suggestion before accepting the completed code.
- Confirm the official Spark availability notice and test that the chosen current alternative can create a new sample app.
- Open a supported test file of the affected language and confirm suggestions appear after the setting change.
- Open a permitted source file and verify Copilot is enabled and generates a test inline suggestion.
Explanation
Correct Answer
B:Confirm the official Spark availability notice and test that the chosen current alternative can create a new sample app.
Incorrect Answers
A:Begin a descriptive function and inspect the ghost-text suggestion before accepting the completed code. Verification of this result involves confirming the official Spark availability notice and test that the chosen current alternative can create a new sample app.
C:Generated code requires independently defined acceptance tests and review against the real program contract. Verification of this result involves confirming the official Spark availability notice and test that the chosen current alternative can create a new sample app.
D:Generated code requires independently defined acceptance tests and review against the real program contract. Verification of this result involves confirming the official Spark availability notice and test that the chosen current alternative can create a new sample app. The observed condition is a team has implemented the following change: Follow the current GitHub Spark lifecycle notice: do not plan new Spark apps, and choose a supported alternative.
Question 197
During troubleshooting, investigators observe: The new subscriber cannot create the tutorial’s Spark application even with a valid Copilot subscription. The system context is: A new Copilot subscriber follows a dated training exercise that asks them to create a GitHub Spark app in October 2026, but the create-app workflow is unavailable. Which ONE cause most directly explains the observed failure?
- The local editor integration and account authentication were not configured.
- GitHub Spark stopped accepting new users and new application creation on August 4, 2026.
- Inline completions were not used for an in-place continuation task.
- A language-specific editor setting overrides the general feature enablement.
Explanation
Correct Answer
B:The remedy should therefore be: Follow the current GitHub Spark lifecycle notice: do not plan new Spark apps, and choose a supported alternative prototyping workflow.
Incorrect Answers
A:The local editor integration and account authentication were not configured. The missing change involves gitHub Spark stopped accepting new users and new application creation on August 4, 2026.
C:Inline completions were not used for an in-place continuation task would not address the need for gitHub Spark stopped accepting new users and new application creation on August 4, 2026.
D:A language-specific editor setting overrides the general feature enablement would not address the need for gitHub Spark stopped accepting new users and new application creation on August 4, 2026.
Question 198
The current design is: Follow the current GitHub Spark lifecycle notice: do not plan new Spark apps, and choose a supported alternative prototyping workflow. The prior failure was: The new subscriber cannot create the tutorial’s Spark application even with a valid Copilot subscription. Which ONE ongoing control best protects against recurrence of that specific failure?
- Review accepted inline suggestions and test their semantics before committing.
- Check per-language settings before changing organization-level Copilot policy.
- Retire legacy Spark onboarding instructions and check first-party feature lifecycle notices before publishing training steps.
- Audit IDE plugin versions and authenticated account after new workstation provisioning.
Explanation
Correct Answer
C:GitHub Spark stopped accepting new users and new application creation on August 4, 2026. Retire legacy Spark onboarding instructions and check first-party feature lifecycle notices before publishing training steps.
Incorrect Answers
A:AI-assisted postmortem claims need links to actual commits, logs and reproducible failures. It would leave retire legacy Spark onboarding instructions and check first-party feature lifecycle notices before publishing training steps without the specified safeguard.
B:Generated code requires independently defined acceptance tests and review against the real program contract. It would leave retire legacy Spark onboarding instructions and check first-party feature lifecycle notices before publishing training steps without the specified safeguard.
D:Audit IDE plugin versions and authenticated account after new workstation provisioning. It would leave retire legacy Spark onboarding instructions and check first-party feature lifecycle notices before publishing training steps without the specified safeguard.
Question 199
A team reports: An author wants Copilot to summarize a 40-file pull request for reviewers who lack time to read every commit individually. They observe: The summary says no API contract changed despite a breaking response-field rename. Which ONE technical decision most directly corrects the cause?
- Start an authorized cloud agent task scoped to the repository and base branch, then review its proposed pull request.
- Reference the approved file with the CLI's supported `@path` context syntax.
- Provide supported repository and path-scoped review instructions and verify the policy is enabled for the repository.
- Generate an AI-assisted pull request summary, then verify its statements against the actual changed files.
Explanation
Correct Answer
D:A generated change narrative can omit important modifications.
Incorrect Answers
A:Start an authorized cloud agent task scoped to the repository would not generate an AI-assisted pull request summary, then verify its statements against the actual changed files.
B:Reference the approved file with the CLI's supported `@path` context syntax would not generate an AI-assisted pull request summary, then verify its statements against the actual changed files.
C:Providing supported repository and path-scoped review instructions and verifying the policy is enabled for the repository would not generate an AI-assisted pull request summary, then verify its statements against the actual changed files.
Question 200
A team has implemented the following change: Generate an AI-assisted pull request summary, then verify its statements against the actual changed files. The original requirement was: An author wants Copilot to summarize a 40-file pull request for reviewers who lack time to read every commit individually. Which ONE validation test best demonstrates the intended behavior?
- Open a test pull request containing an intentionally unsafe access check and inspect the AI review and human gate.
- Sample claims from the summary and confirm the corresponding diff sections and tests exist.
- Verify the draft branch, changed files, checks and required pull-request approvals before merging.
- Ask a factual question about a specific section and verify the file reference was used.
Explanation
Correct Answer
B:Sample claims from the summary and confirm the corresponding diff sections and tests exist.
Incorrect Answers
A:Generated code requires independently defined acceptance tests and review against the real program contract. Verification of this result involves sample claims from the summary and confirm the corresponding diff sections and tests exist.
C:Verify the draft branch, changed files, checks and required pull-request approvals before merging. That check cannot demonstrate tested AI-assisted changes that preserve stated requirements.
D:Ask a factual question about a specific section and verify the file reference was used. Verification of this result involves sample claims from the summary and confirm the corresponding diff sections and tests exist.
GitHub Copilot (GH-300) Practice Tests