1
WGU D685 Practical Applications of Prompt
OA and PA Exam with 250 Questions Actual
Exam 2026/2027 – Complete Exam-Style
Questions with Detailed Rationales | Pass
Guaranteed – A+ Graded
[SECTION 1: Fundamentals of Prompt Engineering — Questions 1-40]
Q1: Which of the following best defines "Prompt Engineering"?
A. The process of writing software code to train a Large Language Model (LLM).
B. The design and refinement of input prompts to elicit desired responses from AI language
models. [CORRECT]
C. The hardware engineering required to cool GPU clusters used for AI.
D. The database management system used to store training data.
Correct Answer: B
Rationale: Prompt Engineering is the art and science of crafting inputs (prompts) that guide an
AI model to produce the most relevant and accurate output. It does not involve coding the model
itself (A) or hardware engineering (C). Option D describes data infrastructure, not prompt
interaction.
Q2: In the context of prompt structure, what is the "Instruction" component?
A. The specific data the model needs to process.
B. The background information setting the scene.
C. The task description that tells the model what to do. [CORRECT]
D. The format in which the answer should be presented.
Correct Answer: C
,2
Rationale: The instruction is the core command (e.g., "Summarize this text," "Translate this
sentence"). It directs the model's action. Option A refers to Input Data, B refers to Context, and D
refers to Output Format.
Q3: Which of the following is an example of "Zero-Shot Prompting"?
A. "Translate 'cat' to French. Dog: Chien. Cat: ?"
B. "Translate 'cat' to French." [CORRECT]
C. "Translate 'cat' to French. Use the suffix '-re' for feminine nouns."
D. "Translate 'cat' to French. Previous translation: Chat."
Correct Answer: B
Rationale: Zero-shot prompting involves giving the model a task without any examples. Option
B provides the task and the input only. Options A and D provide examples (few-shot), and C
provides a specific rule or hint (one-shot).
Q4: What is the primary purpose of "Few-Shot Prompting"?
A. To increase the response length.
B. To provide examples that guide the model toward the desired output format and style.
[CORRECT]
C. To reduce the cost of the API call.
D. To allow the model to forget its training data.
Correct Answer: B
Rationale: Few-shot prompting involves giving the model examples (shots) within the prompt.
This in-context learning helps the model understand the pattern, tone, and format expected. It
does not inherently reduce cost (C) or erase training data (D).
Q5: Which component of a prompt is responsible for providing background information or
setting the scene?
A. Output Indicator
,3
B. Context [CORRECT]
C. Instruction
D. Input Data
Correct Answer: B
Rationale: Context provides the necessary background, persona, or previous conversation history
that helps the model understand the situation. Instruction (C) is the command, and Input Data (D)
is the specific content to act upon.
Q6: What is the "Output Format" in a prompt?
A. The language the model speaks.
B. The speed of the response.
C. The structure in which the response should be presented (e.g., JSON, bullet points).
[CORRECT]
D. The token limit of the response.
Correct Answer: C
Rationale: The output format explicitly tells the model how to structure its answer, such as
"Output in JSON format" or "Provide a bulleted list." This ensures the response is machine-
readable or easy to scan.
Q7: Which technique encourages the model to "think step-by-step" to solve complex logic
problems?
A. Role-Based Prompting
B. Chain-of-Thought (CoT) Prompting [CORRECT]
C. Zero-Shot Prompting
D. Prompt Chaining
Correct Answer: B
, 4
Rationale: Chain-of-Thought prompting instructs the model to break down a problem into
intermediate steps. This improves accuracy on reasoning tasks (like math or logic) by preventing
the model from jumping to a conclusion. Role-Based (A) sets a persona, not a reasoning process.
Q8: What is "Context Window" in relation to LLMs?
A. The time limit for processing a prompt.
B. The maximum number of tokens the model can consider at one time. [CORRECT]
C. The graphical interface used to chat with the AI.
D. The database of previous user inputs.
Correct Answer: B
Rationale: The context window represents the limit of text (measured in tokens) the model can
"remember" or process in a single forward pass. Anything outside this window is ignored. It is
not a time limit (A) or a GUI (C).
Q9: In prompt design, what is a "Constraint"?
A. The data the model processes.
B. The examples provided in the prompt.
C. A limitation or rule imposed on the output (e.g., "limit to 100 words"). [CORRECT]
D. The instruction to ignore previous inputs.
Correct Answer: C
Rationale: Constraints define boundaries for the model's response, such as length, tone (be
formal), or content (do not use technical jargon). Option A is Input Data, B is Few-Shot
examples.
Q10: Which of the following is a primary benefit of using "System Messages"?
A. They allow the user to change the model's training data.
B. They set the high-level behavior, rules, and persona for the AI assistant. [CORRECT]
WGU D685 Practical Applications of Prompt
OA and PA Exam with 250 Questions Actual
Exam 2026/2027 – Complete Exam-Style
Questions with Detailed Rationales | Pass
Guaranteed – A+ Graded
[SECTION 1: Fundamentals of Prompt Engineering — Questions 1-40]
Q1: Which of the following best defines "Prompt Engineering"?
A. The process of writing software code to train a Large Language Model (LLM).
B. The design and refinement of input prompts to elicit desired responses from AI language
models. [CORRECT]
C. The hardware engineering required to cool GPU clusters used for AI.
D. The database management system used to store training data.
Correct Answer: B
Rationale: Prompt Engineering is the art and science of crafting inputs (prompts) that guide an
AI model to produce the most relevant and accurate output. It does not involve coding the model
itself (A) or hardware engineering (C). Option D describes data infrastructure, not prompt
interaction.
Q2: In the context of prompt structure, what is the "Instruction" component?
A. The specific data the model needs to process.
B. The background information setting the scene.
C. The task description that tells the model what to do. [CORRECT]
D. The format in which the answer should be presented.
Correct Answer: C
,2
Rationale: The instruction is the core command (e.g., "Summarize this text," "Translate this
sentence"). It directs the model's action. Option A refers to Input Data, B refers to Context, and D
refers to Output Format.
Q3: Which of the following is an example of "Zero-Shot Prompting"?
A. "Translate 'cat' to French. Dog: Chien. Cat: ?"
B. "Translate 'cat' to French." [CORRECT]
C. "Translate 'cat' to French. Use the suffix '-re' for feminine nouns."
D. "Translate 'cat' to French. Previous translation: Chat."
Correct Answer: B
Rationale: Zero-shot prompting involves giving the model a task without any examples. Option
B provides the task and the input only. Options A and D provide examples (few-shot), and C
provides a specific rule or hint (one-shot).
Q4: What is the primary purpose of "Few-Shot Prompting"?
A. To increase the response length.
B. To provide examples that guide the model toward the desired output format and style.
[CORRECT]
C. To reduce the cost of the API call.
D. To allow the model to forget its training data.
Correct Answer: B
Rationale: Few-shot prompting involves giving the model examples (shots) within the prompt.
This in-context learning helps the model understand the pattern, tone, and format expected. It
does not inherently reduce cost (C) or erase training data (D).
Q5: Which component of a prompt is responsible for providing background information or
setting the scene?
A. Output Indicator
,3
B. Context [CORRECT]
C. Instruction
D. Input Data
Correct Answer: B
Rationale: Context provides the necessary background, persona, or previous conversation history
that helps the model understand the situation. Instruction (C) is the command, and Input Data (D)
is the specific content to act upon.
Q6: What is the "Output Format" in a prompt?
A. The language the model speaks.
B. The speed of the response.
C. The structure in which the response should be presented (e.g., JSON, bullet points).
[CORRECT]
D. The token limit of the response.
Correct Answer: C
Rationale: The output format explicitly tells the model how to structure its answer, such as
"Output in JSON format" or "Provide a bulleted list." This ensures the response is machine-
readable or easy to scan.
Q7: Which technique encourages the model to "think step-by-step" to solve complex logic
problems?
A. Role-Based Prompting
B. Chain-of-Thought (CoT) Prompting [CORRECT]
C. Zero-Shot Prompting
D. Prompt Chaining
Correct Answer: B
, 4
Rationale: Chain-of-Thought prompting instructs the model to break down a problem into
intermediate steps. This improves accuracy on reasoning tasks (like math or logic) by preventing
the model from jumping to a conclusion. Role-Based (A) sets a persona, not a reasoning process.
Q8: What is "Context Window" in relation to LLMs?
A. The time limit for processing a prompt.
B. The maximum number of tokens the model can consider at one time. [CORRECT]
C. The graphical interface used to chat with the AI.
D. The database of previous user inputs.
Correct Answer: B
Rationale: The context window represents the limit of text (measured in tokens) the model can
"remember" or process in a single forward pass. Anything outside this window is ignored. It is
not a time limit (A) or a GUI (C).
Q9: In prompt design, what is a "Constraint"?
A. The data the model processes.
B. The examples provided in the prompt.
C. A limitation or rule imposed on the output (e.g., "limit to 100 words"). [CORRECT]
D. The instruction to ignore previous inputs.
Correct Answer: C
Rationale: Constraints define boundaries for the model's response, such as length, tone (be
formal), or content (do not use technical jargon). Option A is Input Data, B is Few-Shot
examples.
Q10: Which of the following is a primary benefit of using "System Messages"?
A. They allow the user to change the model's training data.
B. They set the high-level behavior, rules, and persona for the AI assistant. [CORRECT]
