Saturday, September 19, 2026

Building Secure Azure AI Foundry Agents with Managed Identity

 

The credential problem nobody wants to own

Every team that connects an Azure AI Foundry agent to Microsoft Graph, SharePoint, Outlook, or a custom API eventually hits the same question: where do the credentials live?

The usual answers are client secrets, certificates, or a service account password tucked into a config file. All three work. All three also mean somebody now owns a rotation schedule, a secret store, and a risk register entry.

Managed Identity removes the question entirely. Azure gets the token; your agent never sees a secret.

This post walks through configuring an Azure AI Foundry agent to call Microsoft Graph using Managed Identity, including the permission model, the scoping controls, and the three errors you will almost certainly hit on the way.

Why Managed Identity wins

Managed Identity lets an Azure resource request Entra ID tokens on its own behalf. Nothing is stored, nothing expires on your watch.

What you get:

  • No secrets in application code or configuration
  • Token acquisition and renewal handled by the platform
  • A smaller attack surface and a cleaner compliance story
  • Native support for least-privilege access
  • First-class integration across Azure services

What you give up: the illusion that permissions come for free. More on that below.

The request path

Azure AI Foundry Agent
        │
        ▼
  Managed Identity
        │
        ▼
  Microsoft Entra ID
        │
        ▼
   Access Token
        │
        ▼
 Microsoft Graph API
        │
        ▼
Exchange Online / Teams / Users / Files

Step 1: Create a user-assigned managed identity

In the Azure Portal, go to Managed Identities → Create and supply a name, subscription, resource group, and region.

mrclmedpggblweuaip-mi

Once it exists, record three values. You will need all of them later:

  • Client ID
  • Object ID
  • Resource ID

User-assigned identities are the right default for enterprise work. They survive resource deletion, can be shared across services, and give you a stable object to audit.

Step 2: Assign Microsoft Graph permissions

This is the step most people skip, and it is the reason most first attempts fail.

A managed identity is an identity, not an authorization. Creating one grants access to exactly nothing. Microsoft Graph application permissions have to be assigned to it separately, as app role assignments on the Graph service principal.

PermissionPurpose
User.Read.AllSearch and read directory users
Calendars.ReadRead mailbox calendars
Mail.ReadRead mailbox messages
OnlineMeetings.Read.AllRead Teams meeting details
OnlineMeetingTranscript.Read.AllRead meeting transcripts

Then grant admin consent. Without it, every call returns:

Authorization_RequestDenied

Assign only what the agent actually needs. An agent that reads calendars has no business holding Mail.Read.

Step 3: Scope mailbox access (optional but recommended)

Application permissions on Graph are tenant-wide by default. Mail.Read means every mailbox in the organization.

For most enterprises that is far too broad, particularly during a pilot. Application Access Policies in Exchange Online restrict a given application to a specific mail-enabled security group.

A typical pilot limits Calendars.Read and Mail.Read to a small group of approved test users. Any mailbox outside that group returns:

json
{
  "error": {
    "code": "ErrorAccessDenied"
  }
}

Worth noting for troubleshooting: this is a different failure from a missing permission, and it looks different in the response. More on that in the error section.

Step 4: Create the Foundry agent

In Azure AI Foundry, open your project and go to Agents → Create Agent.

Configure the model (GPT-5.6 or whatever your organization has approved), the instructions, the tools, and any knowledge sources.

The instructions matter more than they look. Tell the agent to call tools rather than answer from its own knowledge:

You are an enterprise assistant.

Use the available tools to retrieve user, calendar, email and meeting
information whenever it is required. Always call the appropriate tool
rather than generating the information yourself.

Step 5: Add an OpenAPI tool

Go to Agent → Tools → Add Tool → OpenAPI and set the authentication:

Authentication Method : Managed Identity
Audience              : https://graph.microsoft.com

The audience is what tells Foundry which resource to request a token for. Get it wrong and you will get a 401 that looks like a permissions problem but is not.

Step 6: Start with a diagnostic schema

Resist the urge to paste in your full business schema on the first pass. Register one endpoint, confirm the pipe works end to end, then build on it.

A good first probe:

http
GET /users/{userEmail}/calendar

A successful call confirms three things at once: the managed identity acquired a token, Graph accepted it, and Calendars.Read is present and consented.

Two more worth running:

http
GET /users?$top=1

Validates User.Read.All.

http
GET /users/{userEmail}/messages?$top=1

Validates Mail.Read.

Run each from the Foundry playground and inspect Traces → Tool Execution afterwards. The trace shows the resolved URL, the audience, and the status code, which is where the actual diagnosis happens.

Three errors and what they actually mean

Authorization_RequestDenied (403)

The Graph application permission is missing, or admin consent was never granted.

Check the app role assignment on the managed identity, then check consent. These are two separate things and the second is easy to forget.

ErrorAccessDenied (403)

The permission exists, but the mailbox is outside the Application Access Policy scope.

This is the friendlier failure. Authentication worked and the permission is in place; the policy simply blocked this particular mailbox. Verify the mailbox is in the allowed group.

401 Unauthorized

Authentication itself failed. Usually one of:

  • The managed identity is not assigned to the Foundry resource
  • The audience URL is wrong
  • Token acquisition failed

Nothing about Graph permissions will fix a 401. Work backwards from the identity assignment.

The distinction between the two 403s is the single most useful thing to internalize here. One means "you were never allowed to ask." The other means "you may ask, but not about this mailbox."

Practices worth adopting

Prefer user-assigned identities for anything that will outlive a single resource or be shared across services.

Grant the minimum. Every Graph application permission is tenant-wide until you scope it.

Scope mailboxes with Application Access Policies, especially during pilots. Tenant-wide Mail.Read in a test environment is a finding waiting to happen.

Validate with diagnostic tools before writing business logic. A one-endpoint probe tells you whether the authentication chain works. A full schema tells you nothing until it does.

Document the permission-to-API mapping. When an auditor asks why this identity holds Mail.Read, you want a table, not a recollection.

Avoid client secrets wherever managed identity is supported.

Closing

Managed Identity is the right default for Foundry agents that talk to Microsoft Graph and other enterprise services. It removes secrets from the equation while leaving authentication and authorization where they belong, in Entra ID.

Pairing Foundry agents, managed identity, scoped Graph permissions, and Application Access Policies gives you agents that are useful without being over-privileged, and a governance story you can actually defend.

The setup takes an afternoon. The alternative takes a rotation schedule.

Monday, March 16, 2026

🚀 Building a Copilot‑based Meeting Scheduler (POC) using Microsoft Graph & Power Automate

 A practical, step‑by‑step guide for the Microsoft community


📌 Why this article?

Many teams are exploring Copilot Studio to build conversational experiences like:

“Can you help me find a good time for a meeting with my team?”

Behind the scenes, this requires:

  • Checking user availability
  • Respecting working hours
  • Suggesting realistic time slots

This is where Microsoft Graph shines — especially the findMeetingTimes API.

However, when people try this for the first time, they often hit challenges:

  • Premium license errors
  • Graph connection confusion
  • Copilot + Power Automate integration doubts

This article walks through a clean Proof‑of‑Concept (POC) approach, keeping things simple, understandable, and production‑ready later.


🧠 What we are building (POC scope)

✅ Copilot asks for meeting details
✅ Power Automate calls Microsoft Graph
✅ Graph suggests best available time slots
✅ Copilot responds with suggestions

🚫 No auto‑booking (yet)
🚫 No heavy governance (POC only)


🏗️ High‑level Architecture (POC)

User
 ↓
Copilot Studio Agent
 ↓
Power Automate Flow (POC)
 ↓
HTTP with Microsoft Entra ID
 ↓
Microsoft Graph – findMeetingTimes
 ↓
Suggested time slots
 ↓
Copilot responds to user

📎 Screenshot idea:

Diagram slide or whiteboard view of Copilot → Flow → Graph


🧩 Key Components Explained (in simple terms)

1️⃣ Copilot Studio

This is the conversation layer.
It collects:

  • Attendees
  • Meeting duration
  • Preferred date/time range

Copilot does not talk to Graph directly — it calls Power Automate.

📎 Screenshot: Copilot topic with trigger phrases and variables


2️⃣ Power Automate (Heart of the POC)

Power Automate acts as the orchestrator:

  • Receives input from Copilot
  • Calls Microsoft Graph
  • Formats the response back to Copilot

👉 For POC, the Power Automate Premium Trial is sufficient.

📎 Screenshot: Cloud flow triggered by Copilot


3️⃣ Microsoft Graph – findMeetingTimes

This API:

  • Looks at calendars
  • Considers working hours
  • Suggests realistic meeting slots

This avoids brute‑force checking and handles all the logic for you.


🔐 POC Licensing – Keep It Simple

For POC:

  • ✅ Start Power Automate Premium trial
  • ✅ Use your own user account
  • ✅ No service account needed yet

⚠️ Microsoft Graph in Power Automate is a Premium connector — this is expected and by design.

📎 Screenshot: Power Automate “Try premium” banner


🔗 Configuring Microsoft Graph Connection (POC)

Recommended connector

HTTP with Microsoft Entra ID

Why?

  • Simple
  • Supported
  • No custom connector maintenance

Connection setup

FieldValue
Base Resource URLhttps://graph.microsoft.com
Entra ID Resource URIhttps://graph.microsoft.com
AuthenticationUser sign‑in (POC)

📎 Screenshot: Creating HTTP with Entra ID connection



Suggestions

  1. Pick top 3 slots
  2. Convert to friendly text:
    • “Tuesday 11:00–11:30”
    • “Wednesday 3:00–3:30”

Copilot then replies conversationally:

“I found a few good options for everyone. Would you like to go with Tuesday at 11 AM?”


✅ Why this POC approach works well

✔ Minimal setup
✔ Matches real‑world behaviour
✔ No reinventing scheduling logic
✔ Easy upgrade path to PROD


🔜 How this POC evolves into PROD (Later)

POCProduction
User accountService account
Premium trialPer‑Flow license
Suggest onlyAuto‑book meetings
Light governanceFull security & auditing

Graph payloads stay the same
Copilot logic stays the same

Only ownership & licensing change.


❗ Common Pitfalls (and how to avoid them)

  • ❌ Trying to avoid Premium connectors → Not supported
  • ❌ Direct Graph calls from Copilot → Not possible
  • ❌ Using shared mailbox → Unsupported
  • ✅ Always go through Power Automate

📝 Final Thoughts

If you’re exploring Copilot Studio + scheduling scenarios, start small:

  • Prove value with findMeetingTimes
  • Keep the conversation human
  • Let Microsoft Graph do the heavy lifting

Once stakeholders see the value, moving to a production‑grade model is straightforward.


🙋‍♂️ I’d love community feedback!

If you’ve tried similar Copilot scheduling ideas or have tips, please comment and share your experience.
Happy to iterate and learn together 🚀

Thursday, November 6, 2025

Why Power Automate Premium with AI Builder is a Game-Changer for Single Users

 Introduction

Manual data processing is not only time-consuming but also prone to errors, impacting efficiency and decision-making. Organisations need smarter solutions to stay competitive. Enter Power Automate Premium with AI Builder—a cost-effective automation tool that delivers rapid ROI and advanced AI capabilities.


The Problem

  • Manual workflows slow down operations.
  • High error rates lead to rework and inefficiencies.
  • Limited flexibility in customising user interfaces.

The Solution

Power Automate Premium with AI Builder empowers businesses to automate repetitive tasks and integrate AI-driven insights seamlessly. For a single user, the annual cost is just $180, making it an affordable entry point into intelligent automation.


Key Benefits

  • Lower Initial Investment: $180/year vs costly manual labour.
  • Faster ROI: Achieve returns in 6–8 weeks, compared to 3–4 months.
  • High ROI Percentage: 98.5%, ensuring maximum value.
  • AI Builder Credits: Worth 3.3× the monthly licence cost, enabling advanced automation scenarios.

Cost-Benefit Snapshot

AspectManual ProcessingPower Automate PremiumImprovement
Annual CostHidden labour$180Cost-effective
Processing Time3–4 months6–8 weeksFaster turnaround
Error RateHighLowImproved accuracy

Strategic Impact

  • Reduces manual errors and improves data accuracy.
  • Accelerates process efficiency and decision-making.
  • Positions the organisation for scalable automation and AI adoption.



Conclusion

For businesses seeking quick wins and long-term efficiency, Power Automate Premium with AI Builder offers unmatched value. With minimal investment and rapid ROI, it’s the perfect starting point for intelligent automation.

Copilot Studio Bot Error – “This Agent is Currently Unavailable. It Has Reached Its Usage Limit” – Root Cause, Resolution & Best Practices

 Overview

Many organizations have recently reported an issue where users interacting with Microsoft Copilot Studio bots receive the following message:

“This agent is currently unavailable. It has reached its usage limit. Please try again later.”

This can block user conversations with bots in production or POC scenarios.

This is a known issue and can be tracked here:
Power Platform Admin Center Known Issue ID: 5650625


Root Cause Analysis

Behind the scenes, Microsoft introduced a billing enforcement change for Copilot Studio message usage.

Tenants that are using:

  • Viral Trial licenses

  • No paid capacity

  • No paid message packs

  • Or not assigned message capacity correctly

…will hit the message consumption limit and Copilot will stop responding, leading to the above error.

Viral Trial users have very limited message capacity, and when that limit is consumed → Copilot blocks execution.


Resolution / Fix Steps

To fix this for production or active usage scenarios, Microsoft recommends the below options:


Can Credits Be Assigned at Agent Level Instead of Environment Level?

Yes — Agent-level allocation is possible, but it’s still in Preview.

Reference:
Manage Copilot Studio credits and capacity – Microsoft Learn

Note: Preview features are not recommended for production workloads.

Typically, when you purchase a 25,000 message pack → it applies at tenant level → then you allocate it to specific environments.
If not assigned at bot level → any bot in that environment can consume those messages.


Best Practice / Recommendation

For production-grade Copilot bots → use dedicated production environments.

Do not rely on:

  • Shared default environment

  • Trial environment

  • Dev/Test environment for Production usage

Why? Because:

  • Capacity conflicts

  • Data security concerns

  • Risk of consuming message packs by non-production bots

Reference:
Design your Copilot Studio Production Environment Strategy – Microsoft Learn


Suggested Environment Approach

EnvironmentPurposeCapacity Allocation
DEVBuilding & testingMinimal / none
UATPre-production validationControlled
PRODBusiness users live bot usageDedicated message pack allocation

Summary

This Copilot Studio bot usage limit error is happening due to the new billing enforcement.
Trial capacity is not sufficient for real usage → so message packs must be purchased or PAYG enabled.

Action items:

  • Confirm message consumption limits

  • Allocate message packs to appropriate environments

  • Avoid using preview features in production

  • Separate Dev / UAT / Prod to avoid resource conflicts


Final Thought

As organizations adopt Copilot agents more seriously, capacity management becomes critical IT governance.
Proper licensing and environment planning ensures stable bot performance and avoids unexpected downtime for end users.

Tuesday, September 30, 2025

🧠 Microsoft 365 Copilot Studio Licensing: A Simple Guide

 If you're planning to build your own AI assistants (called copilots) using Microsoft 365 Copilot Studio, it's important to understand how the licensing and pricing work. Here's a simple guide to help you get started.


🔑 Types of Licenses

1. Trial License

  • Free to use
  • Good for testing and learning
  • You can’t publish your copilots with this license

2. User License

  • Also free, but you need to buy Copilot Credits for it to work
  • You can build and publish copilots
  • Must be assigned to users through the Microsoft 365 Admin Center

3. Microsoft 365 Copilot License

  • Comes with your Microsoft 365 Copilot subscription
  • Some copilots (like those used in Teams or SharePoint) don’t use credits

💳 How Pricing Works: Copilot Credits

Copilot Studio uses something called Copilot Credits to measure how much you use the service. Every time your copilot answers a question or performs a task, it uses some credits.

Two Ways to Pay:

1. Prepaid Credit Pack

  • $200 per month for 25,000 credits
  • Best if you know how much you’ll use

2. Pay-As-You-Go (PAYG)

  • $0.01 per credit
  • Billed through your Azure account
  • Good for unexpected or extra usage

💡 Example:

If your copilots use 30,000 credits in a month:

  • First 25,000 credits = covered by the $200 pack
  • Extra 5,000 credits = $50 via PAYG
  • Total = $250

🛠️ How to Manage It

  • Assign licenses in the Microsoft 365 Admin Center
  • Track usage and billing in the Power Platform Admin Center
  • PAYG billing is handled through Azure

✅ Quick Summary

License TypeCostCan Publish?Needs Credits?
TrialFreeNo
User LicenseFreeYes
Microsoft 365 CopilotIncludedSometimes

comparison between Microsoft 365 Copilot and Copilot Studio to help you decide which one suits your needs better:


FeatureMicrosoft 365 CopilotCopilot Studio
PurposeAI assistant inside Microsoft apps (Word, Excel, Outlook, Teams, etc.)Build custom AI agents for business processes
License Cost$30/user/month (add-on to M365 E3/E5/Business)Free user license, but requires Copilot Credits
Who Uses ItEnd users (employees, knowledge workers)Developers, business analysts, automation teams
Use Cases- Drafting emails
- Summarizing meetings
- Creating documents
- Analyzing data
- Automating workflows
- Answering internal queries
- Integrating with APIs and data sources
CustomizabilityLimited to Microsoft appsHighly customizable (flows, plugins, connectors)
Data SourcesMicrosoft Graph (emails, docs, chats)Internal systems, SharePoint, Dataverse, APIs
DeploymentBuilt into Microsoft 365 appsPublish to Teams, websites, Power Apps, etc.
Billing ModelPer-user subscriptionCredit-based (e.g., $200 for 25,000 credits/month)
Best ForEnhancing productivity in daily appsBuilding tailored AI solutions for business processes

🧠 Which One Should You Choose?

  • Choose Microsoft 365 Copilot if you want:

    • AI help inside Word, Excel, Outlook, Teams
    • A productivity boost for everyday tasks
  • Choose Copilot Studio if you want:

    • To build your own AI agents
    • To automate specific business processes
    • To integrate with custom data sources


🚀 Final Thoughts

Microsoft 365 Copilot Studio gives you the tools to build smart AI assistants for your business. With flexible licensing and credit options, you can start small and scale as needed.

Friday, September 19, 2025

GirishKumarMs: 🚀Microsoft Copilot – Free vs. Licensed vs. Copilo...

GirishKumarMs: 🚀Microsoft Copilot – Free vs. Licensed vs. Copilo...: Artificial Intelligence is rapidly transforming workplace productivity. Microsoft’s Copilot ecosystem is at the heart of this shift, but the...

How to Track Enterprise-wide Adoption of Copilot Studio

 When we start using Microsoft Copilot Studio to build chatbots and copilots, one big question comes up:

👉 How do we know if people are actually using it, and if it’s adding value across the whole company?

Copilot Studio does show some basic reports, but if we want to understand adoption across the enterprise, we need to look in a few more places.


1. What You Get by Default in Copilot Studio

Each bot has a built-in dashboard that shows:

  • Conversation results – whether chats were resolved, escalated, abandoned, or unengaged

  • Usage – how often generated answers were useful (good, incomplete, irrelevant)

  • Satisfaction – thumbs up/down reactions and survey responses

This is good for checking one bot, but not enough to understand adoption company-wide.


2. Using Power Platform Admin Center

For a bigger picture, go to:
Power Platform Admin Center → Analytics → Virtual Agents (Copilot Studio)

Here you can see tenant-wide usage across all bots, such as:

  • Total number of sessions

  • Active users across the company

  • Capacity usage (important for licensing and planning)

  • Geographical usage (where people are using the bots most)

This helps with overall health checks and adoption tracking.


3. If Your Bot is in Microsoft Teams

If the bot is deployed in Teams, go to:
Teams Admin Center → Analytics & Reports

You’ll be able to track:

  • How many people are using the bot

  • Whether it’s used more in private chats or team chats

  • Message volume per user or per team

This shows how well the bot is fitting into everyday teamwork and collaboration.


4. What Metrics Matter Most

To see the true value of Copilot Studio across the company, track four areas:

AreaWhat to Measure
AdoptionHow many people use it, how often, and how many come back (retention)
UsageNumber of sessions, top questions/intents, average length of conversations
QualityResolution rate, sentiment, helpful vs. unhelpful answers
ROI (Return on Investment)Time saved, % of automation vs. escalations, overall cost savings

5. Why This Matters

  • Bot-level analytics help you improve one chatbot.

  • Admin Center analytics give you adoption and licensing insights.

  • Teams' analytics show you how well the bots are used in daily work.

If you bring all these together (for example, in a Power BI dashboard), you can clearly show leaders how Copilot Studio is being adopted, how much value it’s creating, and where improvements are needed.


In short:
Copilot Studio is powerful, but you need to look beyond the default reports. Using Power Platform Admin Center and Teams analytics gives you the full story on adoption, quality, and ROI.



Thursday, August 21, 2025

🚀Microsoft Copilot – Free vs. Licensed vs. Copilot Studio (Complete Guide)

Artificial Intelligence is rapidly transforming workplace productivity. Microsoft’s Copilot ecosystem is at the heart of this shift, but there’s often confusion around what’s included in the free trial, what the $30/user/month license unlocks, and what extra value Copilot Studio brings to the table.


📊 Free Copilot Trial vs. Paid Microsoft 365 Copilot

 

Feature / Aspect

Free Copilot Chat (Trial)

Licensed M365 Copilot ($30/user/month)

Cost

Free (with M365 subscription)

$30/user/month (~₹2495 + GST)

Expiry

Limited trial, features may expire

No expiry – licensed with full features

Integration

Standalone chat only

Embedded in Word, Excel, PowerPoint, Outlook, Teams, Loop, Edge

Org Data Access

Manual file uploads

Automatic via Microsoft Graph (emails, docs, chats, SharePoint, OneDrive)

Custom AI / Agents

Not included

Copilot Studio for workflows & AI copilots

Security & Compliance

Basic login only

Enterprise-grade security, governance, auditing

Premium Connectors

Not available

Supported via Power Platform (Salesforce, SAP, ServiceNow, etc.)

Productivity Impact

Limited

Significant – saves 25–37 mins per user/day (trials)

 

🏢 Usage Scenarios Across Departments

• Sales: Pipeline summaries, Salesforce insights, auto-proposals.

• HR: Draft policies, onboarding guides, summarize employee feedback.

• Finance: Automate reports, budget forecasting, Excel reconciliation.

• IT: Incident summaries, knowledge base drafting, ServiceNow integrations.

• General: Meeting notes, email drafting, content creation across apps.


💼 Sales Enablement

• Pull opportunity pipeline data directly into Teams.

• Draft personalized proposals in Word using CRM + Excel inputs.

• Generate PowerPoint decks from long reports or competitor analysis.

• Summarize customer meetings and create action plans instantly.


🔗 Premium Connectors (e.g., Salesforce, SAP, ServiceNow)

• Free trial does not support external systems.
• Licensed Copilot + Copilot Studio enable integration with premium connectors via Power Automate and Power Platform.
• Example:
   - Query Salesforce directly from Teams Copilot.
   - Generate Word/Excel reports from SAP budgets.
   - Create custom AI copilots that connect to ServiceNow tickets.
👉 Note: Premium connectors require additional licensing (per-user or per-flow).


🚫 What’s Not Possible in Free Copilot Trial

• No integration with Word, Excel, Outlook, Teams.

• No access to organization data (manual file uploads only).

• No Copilot Studio or automation workflows.

• No premium connectors like Salesforce, SAP, ServiceNow.

• No enterprise-grade compliance/auditing.

• Limited impact – mostly for Q&A and research pilots.


✅ What Becomes Possible with Licensed Copilot

• Seamless integration across all Microsoft 365 apps.

• Contextual insights via Microsoft Graph (emails, meetings, docs).

• Cross-app intelligence (Word ↔ Excel ↔ Outlook ↔ Teams).

• Governance & analytics for IT and business leaders.

• Copilot Studio access → build workflows, custom copilots, plugins.

• Salesforce/SAP integration via Power Platform connectors.


🔹 Copilot Studio – Going Beyond

• Create custom copilots (e.g., Procurement Copilot, HR Copilot, Finance Copilot).

• Connect external data (SAP, Salesforce, ServiceNow, Workday, APIs).

• Build workflows with Power Automate.

• Accept image/file inputs (e.g., upload invoice → extract details → push to SAP).

• Deploy across Teams, web, mobile, or apps.

• Set role-based access, governance, and analytics.


🏷️ Example – Procurement Copilot

Suppose you create a Procurement Copilot that fetches vendor data from SAP, pulls budget info from Excel, and accesses contract details from SharePoint.

Deployment:
• You can publish it org-wide.
• If only M365 data is used (Excel, SharePoint) → all E3 users can use it.
• If premium connectors (SAP/SFDC) are used → end-users also need premium licenses, or you must enable per-flow licensing.


📌 Licensing Decision Flowchart (Salesforce Example)

• With Salesforce connector → Premium license required.
• Without Salesforce (M365 only) → E3 users OK.


📝 Final Takeaways

• Free Copilot = exploration, pilots, file Q&A.

• Licensed Copilot ($30) = true productivity engine across all M365 apps + enterprise compliance.

• Copilot Studio = design your own copilots, workflows, plugins, and connect to external systems like Salesforce/SAP.

• Leadership Insight: Start with licensed Copilot for productivity. Layer in Copilot Studio to create business-specific copilots (Procurement, Sales, HR, Finance). Use per-flow premium licensing to minimize connector costs for large user bases.


Reference links:

Licensing & Cost Model: Copilot Studio & Microsoft 365 Copilot


Connector & Extensibility Details


Features, Architecture & Deployment


Licensing Context & Governance


Building Secure Azure AI Foundry Agents with Managed Identity

  The credential problem nobody wants to own Every team that connects an Azure AI Foundry agent to Microsoft Graph, SharePoint, Outlook, or...