Agents, Copilot Studio and Business Apps
AB-100 design requires choosing task agents, autonomous agents, prompt and response agents, Copilot Studio topics, fallback, actions, Dynamics 365 copilots, Contact Center channels, and AI in canvas apps.
How to study for Microsoft AB-100
Treat each item as an architecture decision: identify the business process, select the Microsoft AI pattern, then add ALM, telemetry, security, responsible AI, and governance controls.
Core concepts
Concept 1
Agent type should match the business process, autonomy needs, channel, required actions, and human oversight.
Exam cue: Use task agents for bounded work and autonomous agents only when goal-directed operation is justified.
Concept 2
Copilot Studio design includes topics, fallback, prompt actions, agent flows, connectors, reasoning behavior, and channel integration.
Exam cue: Use fallback topics when the agent needs a controlled response outside expected topics.
Concept 3
AI components in Power Apps and Dynamics 365 should fit the app's user experience and business process, not sit apart from it.
Exam cue: Use Contact Center and Dynamics 365 context when customer service or sales workflows are central.
Risk pitfalls and guardrails
Giving an autonomous agent broad actions without business controls.
Guardrail: Avoid defaulting to custom agents, skipping grounding-data checks, or treating prompts and connectors as outside the release process.
Ignoring fallback behavior in Copilot Studio topics.
Guardrail: Avoid defaulting to custom agents, skipping grounding-data checks, or treating prompts and connectors as outside the release process.
Adding AI to a canvas app without changing the surrounding process or data flow.
Guardrail: Avoid defaulting to custom agents, skipping grounding-data checks, or treating prompts and connectors as outside the release process.
Memory anchors
Task Agent
A task agent performs a bounded task with defined inputs, actions, and expected outputs.
Autonomous Agent
An autonomous agent pursues goals across steps and needs stronger guardrails and monitoring.
Prompt and Response Agent
A prompt and response agent generates answers or content from instructions and context.
Copilot Topic
A Copilot Studio topic defines trigger phrases, conversation paths, actions, and responses.
Fallback Topic
A fallback topic handles requests that do not match a designed topic.
Prompt Action
A prompt action uses generative AI to produce or transform content inside a workflow.
Contact Center Channel
A Contact Center channel connects agent behavior to customer interaction channels.
Canvas App AI
Canvas app AI embeds intelligent actions directly into a Power Apps user experience.
Checkpoint rule
Do the check-up only after you can summarize each concept in one sentence and identify one dangerous pitfall from memory.
Knowledge Check (after reading)
Short check-up to confirm understanding of this module.
Check-up Questions
A claims team needs an agent to collect a case number, retrieve status, and explain the next step. It must not change the claim. Which agent design is most suitable?
A replenishment agent should monitor inventory events, investigate shortages, propose purchase orders, and act only within approved supplier and spend limits. Which type best fits?
Answer all questions to submit.
Next step personalized recommendations
Continue learning
Move forward only after this module is stable.
What is Pass Harbor?
Completely free exam prep for 317 U.S. exams.
- Practice questions
- Flashcards
- Study guides
- Mock exams
- No registration
- No paywall
- Start instantly
“No more expensive exam prep. Quality study tools should be accessible to everyone.”