Defining Prompt Expectations for AI Summaries in Canvas Apps

AB-410 By islever Editorial Last updated

Create canvas apps Create prompts and models in AI Hub

Short answer

A The summary should highlight key themes, product features, and recurring feedback.

Question

A company has a product report in PDF format. The report must be rewritten with formal vocabulary and cadence.
The company requires a prompt that can be used within a canvas app to rewrite the report as a Word document. The prompt must reference a business glossary stored in a SharePoint site.
You need to make a suggestion for the expectations portion of the prompt to generate a report summary correctly.
What should you suggest?
  1. The summary should highlight key themes, product features, and recurring feedback. Correct Answer
  2. Use the PDF and the business glossary stored in the SharePoint site.
  3. Format the summary as a list of bullet points, which should be clear and use short sentences.
  4. Create a Microsoft Word document summary.

Key Takeaways

The trap is confusing 'Expectations' (content criteria) with 'Context' (data sources) or 'Output Format' (file type). Expectations dictate what the AI must include in its response.

Learn how to define the expectations portion of an AI prompt in a Canvas App to ensure accurate report summaries using business glossaries. Correct answer is A.

Common mistake: Selecting Option B (referencing data sources) is a common error because it looks like part of the prompt setup, but it belongs in the Context/Input section, not the Expectations section.

Expert Analysis

This question tests the specific syntax and semantic structure required when configuring AI Builder or Copilot Studio prompts within a Canvas App. The core concept is distinguishing between instructions (what to do), context (data sources), and expectations (output format/content constraints).

The question explicitly asks for a suggestion for the expectations portion of the prompt. In AI prompt engineering frameworks used in Microsoft Power Platform, 'Expectations' define how the model should interpret the input and what specific content elements must be included in the output.

  • Option A specifies the content focus: highlighting key themes, product features, and recurring feedback. This directly addresses the 'rewrite' requirement by defining the substantive scope of the summary. It tells the AI what information to prioritize from the PDF and Glossary.
  • Option B describes data sources ('Use the PDF...'), which belongs in the 'Context' or 'Input Data' section, not Expectations.
  • Option C describes formatting ('bullet points', 'short sentences'). While this is part of expectations, it is less critical than the content definition for a 'report summary'. However, often formatting is secondary to content relevance in 'summary' tasks unless specified otherwise. But more importantly, Option A defines the semantic expectation of the summary's value.
  • Option D describes the output format/type ('Create a Word document'). This is an action/instruction, not an expectation of the text content itself. Also, the app generates the summary; the Word doc creation might be a separate step or handled by the platform's export capabilities, but the prompt's job is to generate the text.

The key insight is that 'Expectations' in this context refer to the content criteria for the generated response. Option A provides the most complete definition of what the summary should contain to be useful (themes, features, feedback). Comments support A as the correct answer because it defines the substance of the expectation.

If you selected B, you confused Context with Expectations. If you selected C, you focused on style over substance. If you selected D, you confused the output medium with the content expectation.

The operational consequence of getting this wrong is generating summaries that are either irrelevant (missing key themes) or improperly structured if the AI doesn't know what to prioritize.

Exam Strategy

When analyzing prompt configuration questions, always categorize each component into Input/Context, Instructions, or Output/Expectations. 'Expectations' specifically ask for the qualitative or quantitative criteria of the result (e.g., 'include X, Y, Z'), whereas 'Context' lists the data files or variables available to the model. Focus on the word 'Expectations' to filter out options describing data sources or file formats.

Frequently Asked Questions

What is the difference between Context and Expectations in an AI prompt?

Context provides the data source (e.g., PDF, SharePoint list) that the AI reads. Expectations define the rules for the output, such as what topics to cover, the tone, or specific details to include.

Why is Option D incorrect for the expectations portion?

Creating a Word document is an output action or format, not a content expectation. The prompt's expectation section should define the textual content of the summary, not the container it ends up in.

More AB-410 FAQ →

Discussion 1 contributions

Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.

From public community threads 1

  1. exam_upskill 👍 1 Answered A
    A. The summary should highlight key themes, product features, and recurring feedback.

Reader comments & corrections

No reader corrections yet — spotted an error or have a note? Share it below.

Log in to comment, report an error, or add a note about this question.

Submitted for moderation before publishing. Keep it helpful and respectful.

How to Configure Manager and Position Hierarchy in Power Apps?

S-grade

Configure Business Process Flow Branching and Automation

S-grade

Preventing Table Display Name Changes During Promotion

S-grade

Configuring Visual Insights in Model-Driven Apps

S-grade

← Back to AB-410 Study Guide