πŸ”’ Guided

Pre-launch preview. Authorised access only.

Incorrect code

Guided by A Guide to Cloud
Explore AB-900 AI-901
Guided AB-100 Domain 2
Domain 2 β€” Module 8 of 9 89%
15 of 29 overall

AB-100 Study Guide

Domain 1: Plan AI-Powered Business Solutions

  • Agent Requirements & Data Readiness
  • AI Strategy & the Cloud Adoption Framework
  • Multi-Agent Solution Design
  • Build, Buy, or Extend
  • Generative AI, Knowledge Sources & Prompt Engineering
  • Small Language Models & Model Selection
  • ROI, TCO & Business Case Analysis

Domain 2: Design AI-Powered Business Solutions

  • Copilot in D365 Customer Experience & Service
  • Agent Types: Task, Autonomous & Prompt/Response
  • Foundry Tools & Code-First Solutions
  • Copilot Studio: Topics, Flows & Prompt Actions
  • Power Apps, WAF & Data Processing
  • Extensibility: Custom Models, M365 Agents & Copilot Studio
  • MCP, Computer Use & Agent Behaviours
  • M365 Agents: Teams, SharePoint & Sales/Service in M365 Copilot
  • D365 AI Orchestration: Finance, SCM & Customer Experience

Domain 3: Deploy AI-Powered Business Solutions

  • Agent Monitoring: Tools, Metrics, and Processes
  • Telemetry Interpretation and Agent Tuning
  • Testing Strategy for AI Agents
  • Custom Model Validation and Prompt Best Practices
  • End-to-End Testing for Multi-App AI Solutions
  • ALM Foundations & Data Lifecycle for AI
  • ALM for Copilot Studio Agents
  • ALM for Microsoft Foundry Agents
  • ALM for D365 AI Features
  • Agent Security Free
  • Governance for AI Agents Free
  • Prompt Security & AI Vulnerabilities Free
  • Responsible AI & Audit Trails Free

AB-100 Study Guide

Domain 1: Plan AI-Powered Business Solutions

  • Agent Requirements & Data Readiness
  • AI Strategy & the Cloud Adoption Framework
  • Multi-Agent Solution Design
  • Build, Buy, or Extend
  • Generative AI, Knowledge Sources & Prompt Engineering
  • Small Language Models & Model Selection
  • ROI, TCO & Business Case Analysis

Domain 2: Design AI-Powered Business Solutions

  • Copilot in D365 Customer Experience & Service
  • Agent Types: Task, Autonomous & Prompt/Response
  • Foundry Tools & Code-First Solutions
  • Copilot Studio: Topics, Flows & Prompt Actions
  • Power Apps, WAF & Data Processing
  • Extensibility: Custom Models, M365 Agents & Copilot Studio
  • MCP, Computer Use & Agent Behaviours
  • M365 Agents: Teams, SharePoint & Sales/Service in M365 Copilot
  • D365 AI Orchestration: Finance, SCM & Customer Experience

Domain 3: Deploy AI-Powered Business Solutions

  • Agent Monitoring: Tools, Metrics, and Processes
  • Telemetry Interpretation and Agent Tuning
  • Testing Strategy for AI Agents
  • Custom Model Validation and Prompt Best Practices
  • End-to-End Testing for Multi-App AI Solutions
  • ALM Foundations & Data Lifecycle for AI
  • ALM for Copilot Studio Agents
  • ALM for Microsoft Foundry Agents
  • ALM for D365 AI Features
  • Agent Security Free
  • Governance for AI Agents Free
  • Prompt Security & AI Vulnerabilities Free
  • Responsible AI & Audit Trails Free
Domain 2: Design AI-Powered Business Solutions Premium ⏱ ~15 min read

M365 Agents: Teams, SharePoint & Sales/Service in M365 Copilot

Optimise solution design with agents across Microsoft 365, propose agents for business scenarios, configure Sales and Service in M365 Copilot, and leverage the Power Platform AI hub.

Agents where your users already work

β˜• Simple explanation

The best AI assistant sits at your desk, not in a separate building.

M365 agents live inside the apps people already use every day β€” Teams, SharePoint, Outlook. Instead of asking employees to open a new tool, the AI meets them where they are. A sales rep gets deal insights in Teams. A finance analyst gets report summaries in SharePoint. A customer service agent gets case recommendations right in their workflow.

Sales in Microsoft 365 Copilot (formerly Copilot for Sales) and Service in Microsoft 365 Copilot (formerly Copilot for Service) are purpose-built AI assistants that connect M365 to Dynamics 365, so CRM data flows directly into the M365 experience.

The AI hub in Power Platform is the control centre β€” one place to discover, manage, and govern all AI features across the platform.

The AB-100 exam covers the design of agents across the M365 surface area: Teams (personal, team, and channel agents), SharePoint (site-scoped Q&A, document summarisation), and the broader M365 Copilot extensibility model (declarative agents, message extensions, Copilot connectors).

You also need to orchestrate Sales in M365 Copilot and Service in M365 Copilot β€” understanding their data sources, configuration options, and how they bridge M365 and D365. Finally, the Power Platform AI hub centralises AI governance and discovery.

M365 agent types and extensibility

Agent types and surfaces

Agent TypeWhere It LivesWhat It DoesScope
Declarative agentsM365 Copilot (Teams, Outlook, web)Custom agents with defined instructions, knowledge, and actions β€” the primary extensibility modelPersonal, team, or org-wide
SharePoint agentsSharePoint sitesSite-scoped Q&A agents that answer questions using the site’s document library as knowledgeSite-scoped

Extensibility patterns

PatternWhere It AppliesWhat It DoesScope
Message extensionsTeams conversationsAdd rich cards and actions to Teams messages β€” search, compose, and link unfurlingTeam or org-wide
Copilot plugins and connectorsM365 CopilotExtend what M365 Copilot can do β€” adding skills and data from external systemsOrg-wide (admin deployed)

Agents in Teams

Teams agents operate at three scopes, each with different admin controls:

ScopeWho Can Use ItAdmin ControlBest For
PersonalIndividual userUser installs it themselvesPersonal productivity agents (schedule management, task tracking)
TeamMembers of a specific teamTeam owner managesTeam-specific workflows (project status, team knowledge base)
Org-wideEveryone in the organisationIT admin deploys via admin centreOrganisation-wide agents (HR policies, IT helpdesk, company knowledge)

SharePoint agents

SharePoint agents are scoped to a specific site’s content. They answer questions using the documents, pages, and lists in that site.

Design patterns:

  • Department knowledge base β€” HR site agent answers policy questions from the HR document library
  • Project portal β€” project site agent summarises project status from documents and lists
  • Customer-facing documentation β€” external-facing SharePoint site agent helps partners find technical docs

Sales in M365 Copilot vs Service in M365 Copilot

Sales in M365 Copilot vs Service in M365 Copilot
FeatureAudienceData SourcesKey CapabilitiesD365 Connection
Sales in M365 CopilotSales reps, account managers, sales leadersM365 (emails, meetings, Teams) + CRM (D365 Sales, Salesforce)Email summarisation with deal context, meeting prep with CRM data, opportunity insights, pipeline updates in TeamsReads/writes to D365 Sales β€” syncs contacts, opportunities, activities bidirectionally
Service in M365 CopilotCustomer service agents, support managers, field service techniciansM365 (emails, Teams) + CRM (D365 Customer Service, other CRMs) + knowledge basesCase summarisation, suggested responses from knowledge base, email drafts with case context, real-time agent assist during conversationsReads from D365 Customer Service β€” surfaces case history, knowledge articles, similar case resolutions

Configuration orchestration means planning and sequencing the setup:

  1. Licence assignment β€” assign Sales in M365 Copilot or Service in M365 Copilot licences to the right users
  2. CRM connection β€” connect to D365 (or Salesforce/ServiceNow) and configure data sync
  3. Customisation β€” configure which fields, entities, and views are surfaced in Copilot
  4. Rollout β€” pilot with a small group, gather feedback, adjust configuration, expand
πŸ’‘ Scenario: Adrienne deploys M365 agents across Vanguard

Adrienne Cole (VP Enterprise Tech at Vanguard Financial Group) designs an M365 agent strategy across 5 divisions:

Org-wide declarative agent: β€œVanguard Policy Assistant” β€” answers questions about company policies, compliance requirements, and HR procedures. Knowledge sources: SharePoint sites for HR, Legal, and Compliance. Deployed by IT admin to all 12,000 employees via the Teams admin centre.

SharePoint agents (per division):

  • Investment Banking site β†’ agent answers questions about deal procedures and regulatory filings
  • Wealth Management site β†’ agent helps advisors find product fact sheets and suitability criteria
  • Risk Management site β†’ agent summarises risk reports and regulatory updates

Sales in M365 Copilot: Deployed to the 800 relationship managers in Wealth Management and Commercial Banking. Connected to D365 Sales. Configuration: sync client meetings, surface AUM (assets under management) data in email summaries, and generate prep briefs before client calls.

Service in M365 Copilot: Deployed to the 400 customer service agents in the Contact Centre. Connected to D365 Customer Service. Configuration: case summarisation with compliance context, suggested responses from the knowledge base, and real-time assist during customer calls.

AI hub governance: Yuki Tanaka (compliance) uses the Power Platform AI hub to monitor which AI features are active, who’s using them, and whether any data loss prevention policies are being triggered.

Power Platform AI hub

The AI hub is the central place in Power Platform to discover, manage, and govern AI features:

  • Discover β€” browse available AI capabilities (AI Builder models, Copilot features, prompts)
  • Manage β€” configure AI features, set usage limits, assign access
  • Govern β€” monitor AI usage, enforce DLP policies, review audit logs
  • Measure β€” track adoption, usage patterns, and impact across the organisation
πŸ’‘ Exam tip: Teams agent scopes

The exam tests whether you know the right scope for a Teams agent:

  • β€œOnly I need this agent” β†’ Personal scope
  • β€œMy project team needs this agent” β†’ Team scope
  • β€œAll employees need this agent” β†’ Org-wide scope (requires admin deployment)
  • β€œExternal partners need this agent” β†’ Not Teams β€” use a customer-facing channel (website chat, email)

Scope affects admin control: personal agents are user-managed, team agents are owner-managed, org-wide agents are IT-admin managed. The exam may ask about the governance implications of each scope.

Flashcards

Question

What is the difference between Sales in M365 Copilot and Service in M365 Copilot?

Click or press Enter to reveal answer

Answer

Sales in M365 Copilot is for sales reps β€” it connects M365 to CRM data (D365 Sales or Salesforce) for deal insights, meeting prep, and pipeline updates. Service in M365 Copilot is for support agents β€” it connects M365 to service data (D365 Customer Service) for case summarisation, suggested responses, and real-time agent assist.

Click to flip back

Question

What is a SharePoint agent and what scopes it?

Click or press Enter to reveal answer

Answer

A SharePoint agent is an AI Q&A experience scoped to a specific SharePoint site. It answers questions using the site's documents, pages, and lists as its knowledge base. Scope is always the site β€” it can't access content from other sites unless explicitly configured as knowledge sources.

Click to flip back

Question

What does the Power Platform AI hub do?

Click or press Enter to reveal answer

Answer

The AI hub is the central place in Power Platform to discover available AI capabilities, manage AI feature configuration and access, govern AI usage with DLP policies and audit logs, and measure adoption and impact across the organisation.

Click to flip back

Question

What are the three scopes for Teams agents and who manages each?

Click or press Enter to reveal answer

Answer

Personal scope (user manages β€” installs for themselves), Team scope (team owner manages β€” available to team members), and Org-wide scope (IT admin manages β€” deployed via admin centre to all employees). Scope determines governance: wider scope needs stricter admin control.

Click to flip back

Knowledge check

Knowledge Check

Adrienne wants to give Vanguard's 800 relationship managers AI-powered meeting preparation briefs that include client history from D365 Sales, recent email threads, and upcoming calendar events β€” all surfaced in Teams before each client meeting. Which solution should she propose?

Knowledge Check

A company wants to deploy an AI assistant that answers HR policy questions for all 5,000 employees. The policies are stored in a SharePoint site. The assistant should be available in Teams. What is the recommended deployment approach?

Knowledge Check

Yuki Tanaka (Vanguard compliance) needs to monitor which AI features are active across Power Platform, who is using them, and whether any data loss prevention policies are being triggered. Where should she look?

🎬 Video coming soon

Next up: D365 AI Orchestration: Finance, SCM & Customer Experience β€” orchestrating AI features across D365 apps, extending F&O agent chat with custom knowledge, and mapping features to the right D365 app.

← Previous

MCP, Computer Use & Agent Behaviours

Next β†’

D365 AI Orchestration: Finance, SCM & Customer Experience

Guided

I learn, I simplify, I share.

A Guide to Cloud YouTube Feedback

© 2026 Sutheesh. All rights reserved.

Guided is an independent study resource and is not affiliated with, endorsed by, or officially connected to Microsoft. Microsoft, Azure, and related trademarks are property of Microsoft Corporation. Always verify information against Microsoft Learn.