CertPractice

AB-100 study guide

AB-100: Agentic AI Business Solutions Architect practice questions

Plan, design and deploy agentic AI business solutions with Copilot Studio, Microsoft 365 Copilot, Dynamics 365, Power Platform and Microsoft Foundry. Mapped to the skills measured as of October 14, 2026.

Start free practice Official from Microsoft: Exam page · Study guide (skills measured)

Skills measured

3 skill areas and 74 objectives, as of October 14, 2026. Weights show the share of the exam each area makes up.

Plan AI-powered business solutions (25-30%)
  • 1.1 Assess the use of agents in task automation, data analytics, and decision-making
  • 1.2 Review data for grounding, including accuracy, relevance, timeliness, cleanliness, and availability
  • 1.3 Organize business solution data to be available for other AI systems
  • 1.4 Implement the AI adoption process from the Cloud Adoption Framework for Azure
  • 1.5 Design the strategy for building AI and agents in business solutions
  • 1.6 Design a multi-agent solution by using platforms such as Microsoft 365 Copilot, Copilot Studio, and Microsoft Foundry
  • 1.7 Develop the use cases for prebuilt agents in the solution
  • 1.8 Define the solution rules and constraints when building AI components with Copilot Studio, Microsoft Foundry and Microsoft Foundry Tools
  • 1.9 Determine the use of generative AI and knowledge sources in agents built with Copilot Studio
  • 1.10 Determine when to build custom agents or extend Microsoft 365 Copilot
  • 1.11 Determine when custom AI models should be created
  • 1.12 Provide guidelines for creating a prompt library
  • 1.13 Develop the use cases for customized small language models for the solution
  • 1.14 Provide prompt engineering guidelines and techniques for AI-powered business solutions
  • 1.15 Include the elements of the Microsoft AI Center of Excellence
  • 1.16 Design AI solutions that use multiple Dynamics 365 apps
  • 1.17 Select ROI criteria for AI-powered business solutions, including the total cost of ownership
  • 1.18 Create an ROI analysis for the proposed AI solution for a business process
  • 1.19 Analyze whether to build, buy, or extend AI components for business solutions
  • 1.20 Implement a model router to intelligently route requests to the most suitable model
Design AI-powered business solutions (25-30%)
  • 2.1 Design business terms for Copilot in Dynamics 365 apps for customer experience and service
  • 2.2 Design customizations of Copilot in Dynamics 365 apps for customer experience and service
  • 2.3 Design connectors for Copilot in Dynamics 365 Sales
  • 2.4 Design agents for integration with Dynamics 365 Contact Center channels
  • 2.5 Design task agents
  • 2.6 Design autonomous agents
  • 2.7 Design prompt and response agents
  • 2.8 Propose Microsoft Foundry Tools for a given requirement
  • 2.9 Propose code-first generative pages and the use of an agent feed for apps
  • 2.10 Design topics for Copilot Studio, including fallback
  • 2.11 Design data processing for AI models and grounding
  • 2.12 Design a business process to include AI components in a Power Apps canvas app
  • 2.13 Apply the Microsoft Power Platform Well-Architected Framework to intelligent application workloads
  • 2.14 Determine when to use standard natural language processing, conversational language understanding, or generative AI orchestration in Copilot Studio
  • 2.15 Design agents and agent flows with Copilot Studio
  • 2.16 Design prompt actions in Copilot Studio
  • 2.17 Design AI solutions by using custom models in Microsoft Foundry
  • 2.18 Design agents in Microsoft 365 Copilot
  • 2.19 Design agent extensibility in Copilot Studio
  • 2.20 Design agent extensibility with Model Context Protocol in Copilot Studio
  • 2.21 Design agents to automate tasks in apps and websites by using Computer Use in Copilot Studio
  • 2.22 Design agent behaviors in Copilot Studio, including reasoning and voice mode
  • 2.23 Optimize solution design by using agents in Microsoft 365, including Teams and SharePoint
  • 2.24 Orchestrate AI features in Dynamics 365 apps for finance and supply chain
  • 2.25 Orchestrate AI features in Dynamics 365 apps for customer experience and service
  • 2.26 Propose Microsoft 365 agents for business scenarios
  • 2.27 Orchestrate the configuration of Microsoft 365 Copilot for Sales and Microsoft 365 Copilot for Service
  • 2.28 Propose Microsoft Power Platform AI features, including AI hub
  • 2.29 Design interoperability of the finance and operations agent chats to use additional knowledge sources
  • 2.30 Recommend the process of adding knowledge sources to in-app help and guidance for Dynamics 365 Finance or Dynamics 365 Supply Chain Management apps
Deploy AI-powered business solutions (40-45%)
  • 3.1 Recommend the process and tools required for monitoring agents
  • 3.2 Analyze backlog and user feedback of AI and agent usage
  • 3.3 Apply AI-based tools to analyze and identify issues and perform tuning
  • 3.4 Monitor agent performance and metrics
  • 3.5 Interpret telemetry data for performance and model tuning
  • 3.6 Recommend the process and metrics to test agents
  • 3.7 Create validation criteria of custom AI models
  • 3.8 Validate effective Copilot prompt best practices
  • 3.9 Design end-to-end test scenarios of AI solutions that use multiple Dynamics 365 apps
  • 3.10 Build the strategy for creating test cases by using Copilot
  • 3.11 Design the ALM process for data used in AI models and agents
  • 3.12 Design the ALM process for Copilot Studio agents, connectors, and actions
  • 3.13 Design the ALM process for Microsoft Foundry Agents Service
  • 3.14 Design the ALM process for custom AI models
  • 3.15 Design the ALM process for AI in Dynamics 365 apps for finance and supply chain
  • 3.16 Design the ALM process for AI in Dynamics 365 apps for customer experience and service
  • 3.17 Design security for agents
  • 3.18 Design governance for agents
  • 3.19 Design model security
  • 3.20 Analyze solution and AI vulnerabilities and mitigations, including prompt manipulation
  • 3.21 Review solution for adherence to responsible AI principles
  • 3.22 Validate data residency and movement compliance
  • 3.23 Design access controls on grounding data and model tuning
  • 3.24 Design audit trails for changes to models and data

Sample questions

Sample question 1 · Plan AI-powered business solutions

A finance team spends hours each week copying invoice totals from emails into an ERP system using the same fixed steps. What is the best first candidate for an agent?

  1. A. Negotiating supplier contracts
  2. B. Writing the company's annual report
  3. C. A strategic pricing decision made once a year
  4. D. The repetitive, rules-based invoice data entry task
Show answer and explanation

Answer: D

Agents deliver the clearest value on frequent, repetitive, well-defined tasks with measurable time savings. One-off strategic decisions and negotiations need human judgment and offer little volume to automate.

Learn more: Business plan for AI agents

Sample question 2 · Design AI-powered business solutions

Agents in a contact center use internal abbreviations such as "RMA" and "SLA-P1". Copilot keeps misreading them in case summaries. What should you configure?

  1. A. A larger model
  2. B. Business terms (a glossary) that define the company's terms for Copilot
  3. C. A Power BI dashboard
  4. D. A new security role
Show answer and explanation

Answer: B

Defining business terms with clear descriptions helps Copilot interpret company-specific vocabulary correctly in summaries and answers.

Learn more: Customize Copilot in Dynamics 365 Sales

Sample question 3 · Deploy AI-powered business solutions

Where can makers see built-in metrics for a Copilot Studio agent, such as sessions, resolution rate and escalation rate?

  1. A. The Analytics page in Copilot Studio
  2. B. Windows Event Viewer
  3. C. The Outlook calendar
  4. D. The Power Apps theme editor
Show answer and explanation

Answer: A

Copilot Studio includes analytics for engagement, outcomes, satisfaction and usage, for each agent.

Learn more: Manage AI agents across your organization · Metrics and recommendations for Copilot Studio

Sample question 4 · Plan AI-powered business solutions

An HR agent gives answers based on a policy document that was replaced six months ago. Which data quality dimension failed?

  1. A. Cleanliness
  2. B. Relevance
  3. C. Timeliness
  4. D. Availability
Show answer and explanation

Answer: C

Timeliness means grounding data reflects the current state. Outdated policies lead to confidently wrong answers even when the document is otherwise accurate and relevant.

Learn more: Data architecture for AI agents across your organization

Sample question 5 · Design AI-powered business solutions

A Customer Service manager wants Copilot to use the company's own knowledge articles when drafting answers. What should you configure?

  1. A. Copilot's knowledge sources in Customer Service, pointing to the approved knowledge base
  2. B. A new canvas app
  3. C. A Power BI report
  4. D. An Excel macro
Show answer and explanation

Answer: A

Copilot in Customer Service can be configured to use specific knowledge sources so its answers are grounded in approved company content.

Learn more: Customize Copilot in Dynamics 365 Sales · Configure copilot features for custom case forms and custom apps

Frequently asked questions

How many AB-100 practice questions are there?

246 original questions covering all 74 objectives in Microsoft's skills outline (skills measured as of October 14, 2026). Each one has an explanation, and most link to the relevant Microsoft Learn page.

Is it free?

The first 10 questions are free, with no account needed. Unlocking the full set costs US$9.99 for 30 days of access.

Are these real exam questions?

No. Every question is written from scratch and mapped to the published skills outline. They are not copied from the real exam, so they help you learn the material rather than memorize answers.

How does practice differ from exam mode?

Practice mode lets you check or view the answer after each question. Exam mode is timed (100 minutes), shuffles questions and answers, and scores you at the end.

Practice all AB-100 questions