AB-100: Predict your score.
Know what to study next.
See your predicted score, follow a study plan based on your answers, and revisit topics you're starting to forget. It all runs on your device.
443 practice questions
AI score prediction
100% offline
Download free
iPhone & iPad · Free to startOn your Mac? Scan to install
AB-100Microsoft Agentic AI Architect
AB-100 · 30 days left ›
PREDICTED786 ±37✓ Exam Ready · 87% confidence
AB-100 Practice Questions & Exam Prep — Microsoft Agentic AI Business Solutions Architect
Get exam-ready for AB-100 (Microsoft Agentic AI Business Solutions Architect) on iPhone or iPad. Azure Mastery uses on-device AI to predict your readiness score across all three AB-100 domains — Plan, Design, and Deploy AI-powered business solutions — builds a personalised study plan from your weak spots, and surfaces topics you're forgetting. Core study stays on-device and works offline; optional sync uses your private iCloud account.
The exam
What is the AB-100 exam?
AB-100 is the Microsoft Agentic AI Business Solutions Architect credential — the senior-architect exam for designing, deploying, and governing agentic-first business solutions across the full Microsoft AI stack: Copilot Studio, Microsoft Foundry, Microsoft 365 Copilot, Dynamics 365 Copilot, and Power Platform AI. It covers:
Plan AI-powered business solutions — ROI analysis, multi-agent strategy
Design agents and orchestration — task, autonomous, and prompt-and-response agents, MCP extensibility, Foundry Tools
Deploy with full ALM, telemetry, responsible AI, and security
AB-100 tests architecture decisions across several Microsoft platforms. You'll:
Analyse business requirements and choose whether to build or buy an agent solution
Design multi-agent systems with Agent2Agent (A2A) and Model Context Protocol (MCP)
Work across Microsoft 365 Copilot for Sales and Service and Dynamics 365 AI
Manage lifecycle, governance, and audit trails for agents and models
Scenarios ask you to choose a component, deployment slot, or control that fits the wider business system.
Microsoft's current AB-100 skills are measured as of 22 July 2026. Every question in Azure Mastery's AB-100 bank is mapped to the current outline — including the new MCP extensibility, Computer Use, agent feed, and DSPM-for-AI topics. Read the official outline at learn.microsoft.com.
Questions40–60 questions, mixed formats
Duration100 minutes (120 min seat)
Pass score700 / 1000
CostUSD $165 (≈ £128 UK)
ValidityRenew annually
FormatOnline or test centre
Skills measured · 22 July 2026
AB-100 exam objectives
Three domains from Microsoft's current outline, skills measured as of 22 July 2026. Every domain below lists Microsoft's own skill groups verbatim from the official study guide, so you always know which domain you're being tested on and where your weak spots cluster.
AI & agent skill mapDataModelsAgentsResponsible AI
Plan AI-powered business solutionsPublished weight 25–30%
124 exam-scoped practice questions in the app
Explore Plan topics
Analyze requirements for AI-powered business solutions
Design overall AI strategy for business solutions
Evaluate the costs and benefits of an AI-powered business solution
Covers assessing where agents genuinely help, checking grounding-data quality, and sequencing adoption with the Cloud Adoption Framework's AI track.
The ROI and total-cost-of-ownership case gets skipped in favour of jumping straight to a build-versus-buy answer without the numbers.
Practise arguing when a model router beats a single model on cost versus capability before picking either.
Design AI-powered business solutionsPublished weight 25–30%
131 exam-scoped practice questions in the app
Explore Design topics
Design AI and agents for business solutions
Design extensibility of AI solutions
Orchestrate configuration for prebuilt agents and apps
Spans the most surfaces: task, autonomous, and prompt-and-response agents, plus Microsoft 365 Copilot extensibility choices.
Candidates default to a custom Foundry build when a declarative Microsoft 365 Copilot extension already covers the requirement.
Match Model Context Protocol, Computer Use, or a Copilot Studio topic to the exact extensibility need described.
Deploy AI-powered business solutionsPublished weight 40–45%
188 exam-scoped practice questions in the app
Explore Deploy topics
Analyze, monitor, and tune AI-powered business solutions
Manage the testing of AI-powered business solutions
Design the ALM process for AI-powered business solutions
Design responsible AI, security, governance, risk management, and compliance
The heaviest domain: telemetry interpretation, validation criteria, and lifecycle management across Copilot Studio, Foundry, and custom models.
Governance is the biggest theme here, and candidates underweight prompt-manipulation mitigations and data-residency controls.
Trace one agent's audit trail and grounding-data access controls before assuming security means model safety alone.
Common traps
Where AB-100 candidates slip
Five recurring misconceptions that trip up otherwise well-prepared AB-100 candidates, grounded in the current skills outline.
The right answer matches the stated constraint
Candidates pick a design that's correct for a different constraint than the one the scenario actually states.
Cost, residency, governance, and time-to-market are the four constraints that repeatedly decide the correct answer here.
Read the stem for its stated constraint first, then evaluate every option only against that one requirement.
Smallest sufficient capability beats the most powerful one
Candidates reach for the most powerful platform option when the scenario's requirements are satisfied by a lighter one.
A declarative extension or prebuilt feature beats a custom build whenever it already satisfies the stated requirement.
Ask whether the scenario actually demands the heavier option's extra capability before choosing it.
Know which platform owns a capability by default
Candidates can't say which platform is the intended surface when several technically overlap on the same capability.
Several platforms can technically do the same job, but the exam rewards naming the one Microsoft intends for it.
Decide whether the scenario is extending an existing Copilot experience or building a new agent from scratch.
Responsible AI requirements are design constraints, not an afterthought
Candidates treat responsible AI as a separate checklist item instead of applying the specific safety, privacy, transparency, or oversight requirement in the scenario.
A named Responsible AI requirement can rule out an otherwise viable design, but it is not an automatic tie-breaker when the scenario does not call for that control.
Look for an impact assessment, content-safety classifier, audit trail, or refusal pattern as the tie-breaking detail.
The scenario paragraph carries the answer
Candidates miss a compliance requirement or staffing limit buried in the setup that decides the correct design.
A compliance requirement or staffing limit stated once in the setup often isn't repeated anywhere in the question itself.
Read the full business scenario once before touching any option, including the setup paragraph and the question line.
Designed for AB-100
How Azure Mastery helps you pass AB-100
Exam-specific practice
443 AB-100 practice questions aligned to the published skills outline.
Coverage leans hardest on Deploy, the domain Microsoft weights up to 45%.
Practise matching a design to the one constraint a scenario actually states, not the one that sounds best.
Predicted score
Exam IQ forecasts your AB-100 score on-device after roughly 30 questions, confidence range attached.
It flags whether architecture design or deployment governance is dragging that number down.
Compare Plan, Design, and Deploy side by side to see which one is holding your score back.
Adaptive study plan
Your plan leans hardest on Deploy, the domain Microsoft weights up to 45%.
Miss a governance or lifecycle-management question and the next session surfaces that domain first.
Hold a topic steady across three sessions and the plan moves on to your next weak spot.
Knowledge decay
Forget the smallest-sufficient-capability rule early on and later design questions get harder too.
Watch for a decay flag on the Today screen and revisit that topic before it fades completely.
A Focus Weak Spots session pulls a fading topic back into rotation automatically.
Exam rehearsal
Practise holding focus through a full 100-minute session, matching AB-100's real length.
Each rehearsal session weights Plan, Design, and Deploy the way Microsoft publishes them, using original questions.
The exact interface and question order on exam day are Microsoft's decision, not Azure Mastery's.
Answer Coach
Untangles AB-100's near-identical pairs: which platform owns a capability by default, and when restraint beats capability.
Get a question wrong and Answer Coach names the constraint in the scenario that should have driven your answer.
An optional on-device rewrite is available on supported devices, always checked against the written explanation first.
Aura guidance
Aura adjusts its suggestions as your first week of AB-100 practice takes shape.
A short recap closes every session: what got better, what to revisit, and where to go from here.
Private by design
Your AB-100 answer history, readiness gauge, and Answer Coach notes stay on your device by default.
No Azure Mastery account and no server ever processes your AB-100 study data.
Should you enable sync, it moves through your own private iCloud account and nowhere else.
10-week study plan
Suggested AB-100 study plan
Start with business planning, then design agents and practise deployment. Use this ten-week route at your own pace, with time for testing, governance and review.
Requirements and grounding
Identify automation, analytics and decision-support use cases.
Check grounding data for accuracy, relevance and timeliness.
Organise business data for AI consumption.
AI strategy
Review the Cloud Adoption Framework AI guidance.
Compare Copilot, Copilot Studio and Foundry for multi-agent solutions.
Plan build, buy or extend choices, prompt guidelines and an AI centre of excellence.
Value and investment
Estimate total cost of ownership.
Compare the ROI of building, buying and extending.
Explore AI scenarios across Dynamics 365 apps.
Agent design
Compare task, autonomous and conversational agents.
Design Copilot customisations for Dynamics 365 customer experience and service.
Use Foundry Tools and Copilot Studio topics with fallback.
Extensibility
Explore custom models in Foundry and agents in Microsoft 365 Copilot.
Review MCP extensions and Computer Use.
Design agent reasoning, behaviour and voice interactions.
Prebuilt agents
Explore agents for Dynamics 365 finance, supply chain and customer service.
Compare Microsoft 365 Copilot for Sales and Service scenarios.
Review Power Platform AI Hub and model routing.
Monitoring and tuning
Use agent-monitoring tools and interpret telemetry.
Turn user feedback and backlog findings into improvements.
Review model-performance metrics.
Testing
Define an agent test strategy and success metrics.
Set validation criteria for custom AI models.
Test prompts and end-to-end scenarios across Dynamics 365 apps.
Deployment and governance
Plan ALM for data, agents, models and Dynamics 365 AI.
Review security, access controls and prompt-manipulation mitigations.
Check responsible AI, data residency and audit trails.
Review and rehearse
Use Focus Weak Spots to revisit weaker domains.
Complete two timed, 100-minute Exam Simulator sessions.
Review mistakes and use the results to guide further study.
Inside the app
Nine interactive practice formats, on iPhone
Azure Mastery has nine interactive formats for exam practice. The examples below show original AB-100 practice questions in Azure Mastery on iPhone and iPad. The exam simulator uses timed sessions with the published domain weights. Microsoft's live interface and question mix may differ.
Which Cloud Adoption Framework methodology covers establishing the guardrails an AI workload must operate inside?
Govern
Adopt
Strategy
Ready
Multiple choice
An original AB-100 practice question with one correct answer. The app explains every option after you answer.
Exam-specific sample
Which of the following are core Cloud Adoption Framework methodologies relevant to AI adoption? (Select all that apply.)
Strategy
Plan
Ready
Assess
Manage
Secure
Multi-select
An original AB-100 multi-select question. Select all the correct options to earn the mark.
All-or-nothing
Arrange the application lifecycle management (ALM) stages for promoting a Microsoft Copilot Studio agent solution from development to production, in the…
⋮⋮1Author and test the agent in a development environment
⋮⋮2Export the work as a managed solution
⋮⋮3Import the managed solution into a test environment for validation
⋮⋮4Deploy the validated managed solution to the production environment
Drag-and-drop
An original AB-100 ordering question you can answer by touch on iPhone and iPad.
Interactive item
Select the Cloud Adoption Framework AI stage where the organisation defines motivations, business outcomes, and the operating model for AI before any platform or…
Hotspot
An original AB-100 question with a visual prompt, designed for touch on iPhone and iPad.
Tap target
Contoso Insurance — Claims-Adjuster AI Agent Contoso Insurance is a regulated insurer rolling out an AI agent to help claims adjusters investigate, summarise, and respond to customer claims. The team has Microsoft…
1Which extensibility format should the team choose for the agent?
2Which OAuth flow is appropriate for the agent's downstream API calls?
3Which Responsible AI control set should be baked into the agent's design?
4Which Microsoft surface provides the audit trail the compliance officers require?
Case studies
An original AB-100 case study with several questions about the same requirements and environment.
Multi-question
✕Your answer: Adopt
✨ Answer Coach:Adopt covers the migration and innovation work performed inside guardrails that already exist.
— grounded in authored certification guidance
Answer Coach
Answer Coach explains why the answer is correct and, where available, why each option is right or wrong. It helps you understand a mistake and remember the distinction. Read explanations after each question or at the end of a practice test. On supported devices, optional AI can reword a note on your device after checking it against the written guidance.
App exclusive
Frequently asked
AB-100 FAQs
How much does the AB-100 exam cost?
The AB-100 voucher costs USD $165 in the United States (around £128 in the UK; pricing varies by region). As an Expert-tier exam, AB-100 is rarely included in free-voucher promotions — if one appears around Microsoft Build or Microsoft Ignite, it's worth grabbing. Renewal itself stays free.
Does the AB-100 certification expire?
Yes. Like Microsoft's other role-based certifications, AB-100 expires annually, with a free, short online renewal assessment on Microsoft Learn available in the six-month window before expiry — scoped to recent skills-outline updates. AB-900, the Fundamentals-tier exam, doesn't expire.
What is the AB-100 retake policy if I fail?
Given how broad AB-100's syllabus is, treat a fail as a signal to target the weakest domain rather than rebook immediately.
First retake: after 24 hours.
Second and third retakes: a 14-day wait each.
Cap: five attempts per rolling 12 months, each needing a fresh voucher.
What does the AB-100 exam format look like?
AB-100 runs as a 100-minute session (120 minutes with seat time) of 40–60 questions, online or at a test centre, scored out of 1000 with a 700 pass mark. Azure Mastery's bank leans heavier on multi-paragraph scenarios than the associate-level agent exams, so you read a business constraint before picking the architecture that fits it.
How long should I study for AB-100?
Eight to twelve weeks if you already have hands-on time with at least one of Copilot Studio, Microsoft Foundry, or Dynamics 365 Copilot. Give it three to four months if multi-agent orchestration is genuinely new — AB-100 expects you to design end-to-end solutions rather than describe the individual pieces.
AB-100 vs AB-620 — which agent cert should I take?
AB-620 is for makers who build agents in Copilot Studio. AB-100 is for architects who choose the platform, governance, lifecycle management, and business case across Copilot Studio, Foundry, and Dynamics 365. Start with building agents, then move to AB-100 when your work includes those wider architecture decisions.
AB-100 vs AI-103 vs AB-620 — how do the roles differ?
AI-103 is the code-first developer path (Python, Foundry). AB-620 is the Copilot Studio builder path (low-code agents). AB-100 sits above both as the architecture credential spanning every Microsoft AI surface. Get comfortable with an implementation credential before attempting the architecture-level exam.
Do I need prerequisites before attempting AB-100?
Not formally, but AB-100 assumes it. Microsoft expects candidates to already hold AI-901-level knowledge, plus working experience with at least one of AI-102, AI-103, PL-200, PL-300, PL-400, PL-500, or AB-620. AB-100 tests architectural decisions across platforms you're assumed to already know individually — it isn't where most candidates start their Microsoft AI journey.
Is Azure Mastery free for AB-100 prep?
The app is free to download with a free allowance of AB-100 questions. The full 443-question AB-100 bank unlocks with a one-time exam-pack purchase, or a subscription or lifetime upgrade unlocks every exam in the catalogue.
Does AB-100 practice work offline?
Question practice, scoring, and the readiness prediction all run on-device, so a flight or offline commute won't interrupt your study streak. No account is required, and your data stays on-device unless you turn on optional private iCloud sync.
Free study guides
Free guides that pair with AB-100
Where AB-100 fits
Certification paths that include AB-100
AB-100 sits at the top of Microsoft's agentic AI track — the architect-level credential for solutions that combine Copilot Studio, Microsoft Foundry, Dynamics 365 Copilot, and Power Platform AI. Microsoft recommends entering via AI-901 (AI Fundamentals) and AB-620 (AI Agent Builder) before sitting AB-100. Tap any linked exam below to see its dedicated study app page.
AB-100 is the architecture credential in Microsoft's agentic AI track. AB-620 covers practical Copilot Studio skills, while AI-103 covers development with Foundry and code. Both provide useful experience for the systems you'll design in AB-100.
Ready to pass AB-100?
Download Azure Mastery free. 443 AB-100 practice questions across all three domains, AI score prediction, full-length exam simulator, adaptive study plan. iPhone & iPad.
