Table of Contents
Copilot Studio is Microsoft’s low-code platform for building custom AI agents, in this blog post we will cover the basic concepts to get stated with Copilot Studio and we will cover a simple but powerful use case for an agent that helps users to streamline the preparation process for Monthly Business Review meetings, and the step-by-step on how to build it.
Copilot Studio vs Microsoft Copilot
Before we start building the agent in Copilot Studio, let’s clarify that Copilot Studio and Copilot are different products.
- Copilot (without “Studio” in the name) is the ready-to-use AI assistant built by Microsoft, that can be used in all the apps we already know: Outlook, Excel or PowerPoint. The little icon bugging us everywhere, offering help.
- Copilot Studio is the tool to build our own AI agents. In this blog, we’re referring to Copilot Studio.
What is Copilot Studio?
Microsoft’s official definition for Copilot Studio is as follows:
“Copilot Studio is the AI agent platform for Copilot, where you can build agents that help do work. Effortlessly connect agents to your organization’s data, tools, and systems to multiply capacity and accelerate outcomes. Build agents for any scenario—from simple to sophisticated—using natural language and publishing to where you need them.”
Some key points to note from above definition are:
- Copilot Studio is the tool to build our own “Copilots” or our own “ChatGPTs”
- With Copilot Studio we can build agents that can do work for us
- We can connect agents to data and multiple systems
- Copilot Studio is a Low-Code tool
Let’s create our first agent with Copilot Studio
Some of the examples shown in this blog will be based on Microsoft’s official lab for Copilot Studio: Create a Monthly Business Review Agent.
The purpose is to create an agent that keeps directors and managers informed of the current status of all deals, identifying risks, and aligning key initiatives. The Sales team creates a summary file for each representative that looks as follows:
The files are uploaded to a SharePoint site:
After implementing an agent, the directors will simply ask a question and the agent will scan the latest MBR files in SharePoint and will retrieve a response within seconds.
Create a simple agent
To start with Copilot Studio, go to the portal (https://copilotstudio.microsoft.com/). In the home page, you’ll find a few options to get started.
A simple way to create a quick agent is to “describe” what we’d like the agent to do by using the prompt chat.
We can also use the gear icon to indicate the language and the solution that the agent should belong to.
After using the quick prompt, the agent will be provisioned and both a description and instructions will be added. These can be edited and tuned at any time. On the right side of the screen you’ll be able test the agent in the chat experience.
For now, we have a very simple agent that doesn’t have much context or information to work with, but in the next few sections we will learn more about how to customize it and how to give it information it can answer questions from.
Basic settings
First, let’s talk about some basic but key settings:
Agent's model
Use information from the web
Does your agent need to answer questions using information from the web or just using specific sources? If it only needs to answer or act based on internal sources such as Standard operational procedures (SOPs) or HR policies or a SharePoint site, we need to turn the “use information from the web” option off.
If you think your agent needs to use the web as well, then we can keep it on.
Allow or disable file uploads
We can also define if we want to allow users to upload files
Language(s)
When creating the agent in the gear icon, we can set up the primary language. As of December 2025, the primary language cannot be changed (you can only switch between different regions for “English” but changing from “English” to “Spanish” is not possible) but secondary languages can be added.
Knowledge Source
Now, it’s time for fun! Let’s give the agent some context, i.e. let’s add some source of knowledge. We can do this from the Overview tab or from the Knowledge tab.
As of 2025, the available Knowledge sources are:
- Public Website
- SharePoint site
- Azure AI Search
- Dataverse
- Dynamics 365
- Salesforce
- ServiceNow
- Azure SQL
- Snowflake
- Databricks
- Confluence
- Oracle Database
- SAP OData
- Zendesk
- Azure Databricks
For this simple use case, we are assuming that the files the agent will respond questions from will be in SharePoint:
Now, every time there is a new file added to the SharePoint folder that we indicated, the agent will use it to answer questions Once we added this context the agent can start answering questions about the files in SharePoint:
Tools
Now it’s time to give the agent some additional tasks that go beyond answering questions. We can use tools to have the agent send emails, assign tasks in Planner, Update data in Dataverse, run workflows and much more.
Some examples of tools that can be added are:
- Office 365 Outlook
- Excel
- SharePoint
- Jira
- Planner
- Dataverse
- Model Context Protocol servers
- API connections
Forthe simple agent we’re building, we want to give it the ability to send emails to the directors in case they want to receive the requested information in their inbox. So, let’s add the Office 365 Outlook tool. After selecting the tool, it will ask us which specific actions it should be allowed to perform:
Finally, it will ask us to set up the connection and finalize the set up:
Now, we can start the agent to send the summary through email instead of showing us a summary in the chat.
Publishing the agent
Copilot Studio agents can be deployed and published on many platforms. These can be found in the “Channels” tab.
As of December of 2025, The available channels are:
- Teams
- Microsoft 365 Copilot
- SharePoint
- Website
- Native app
- WhatsApps
- Slack
- Telegram
- Twilio
- Line
- GroupMe
Other Features
Beyond basic conversation design, Copilot Studio also provides a set of advanced features that allow agents to move from simple question-and-answer scenarios to more structured, event-driven, and goal-oriented behaviors.
Triggers
Using Copilot Studio triggers, we can set up different events that will automatically activate an agent when the event happens. Triggers are what converts an agent from conversational to autonomous.
For example, let’s say we have an agent that receives IT support tickets from users in a chat, the agent has been trained with guidelines and policies and works just fine. But let’s assume that we also have an email address (servicedesk@yourcompany.com) where users can send emails asking for help. We can connect the service desk mailbox to the very same agent that was trained for chat, and it will use the very same instructions, tools, data and knowledge and simply act whenever a new request arrives through email.
Some trigger examples are:
- Run on schedule
- When new item is added to SharePoint
- When response is submitted in Microsoft Forms
- When a file is created in a SharePoint site
Does this look familiar? Yes, these are the same triggers we find in Power Automate. Behind the scenes, what triggers do in Copilot Studio is to create a workflow that is called when the event happens (i.e, when the email is received, when the file is created in SharePoint) and run a “prompt” against the agent and all the setup, tools, actions, knowledge and data behind it.
Agent flows
Just like Power Automate triggers can be used from Copilot Studio, regular Power Automate actions and workflows can also be leveraged with Agent Flows.
Agent flows are Power Automate flows that are created from Copilot Studio and they are used to connect agents with automated actions.
Below is an example of a workflow built for a Copilot Studio agent that receives an Order ID from the customer, then queries this order ID in a Dataverse database and updates its status.
Multi-agent orchestration
Copilot Studio allows the creation of multi-agent systems where agents can talk to each other and work together to achieve a shared goal. Instead of having a single agent to do everything, Copilot Studio lets agents can be inter-connected and delegate tasks to one another. This is not only limited to Copilot Studio agents, it can also include agents created in the Microsoft 365 agent builder, Microsoft Azure AI Agents Service and Microsoft Fabric.
Other Copilot Studio use cases
Copilot Studio agents have quite a few business use cases they can help solve. Some of the most scenarios we’ve covered with a few clients are:
IT Service Desk Automation
An agent can have a conversation with a user that is having a technical issue and provide quick fixes based on general guidelines but based on the type, urgency, or content of the request, the agent can direct and escalate it to the appropriate team and create support tickets.
Customer Service Support
Copilot Studio agents can be deployed to a website to allow customers to request updates on their orders or submit payment remittance proofs and have these files attached to the request for further processing.
Leads Management Assistant
Copilot Studio can be leveraged to streamline how sales teams handle incoming leads and information requests. The assistant automatically classifies leads based on intent, priority, and readiness to buy, while routing general inquiries to the appropriate team or workflow.
By adding agent flows and triggers to the agent, it can perform other routine tasks such as status updates, fill out templates and follow-up reminders.
Turn AI into value with Copilot Studio
Whether you have some ideas in mind or if you’d just like to have a conversation to learn what we’ve done on Copilot Studio and other Artificial Intelligence integration with all the Microsoft 365 ecosystem, contact us, we’ll be more than happy to have a conversation and see how we can be of help for your team.
