This might be the week the bots stop “assisting”… and start working.
Microsoft quietly flipped a switch — and Copilot Studio can now literally use your computer.
Not API calls. Not connectors. Not cloud sandboxes.
Actual mouse movement. Real keyboard input.
A legit AI agent that can launch your Power App, fill the fields, and submit the form — like a disturbingly compliant intern.
In this episode we unpack the feature Microsoft calls Computer Use — the update that turns Copilot into a hands-on operator of Windows machines. We walk through setup, the security ceremony no one warns you about, and then watch the AI stumble, misclick, recover, adapt… and eventually succeed. It’s messy, slow, hilarious — and also historic.
This is agentic AI in the enterprise — the moment automation stops being a diagram and becomes a digital worker.
If your business runs on legacy apps, intranet buttons, and “almost integrated” everything… this is the episode you need to hear. Because this is where Copilot stops writing emails — and starts doing tasks.
You will either fear for your job… or go rewrite your job title to “AI Wrangler.”
You can now Control My Power App more easily thanks to Microsoft Copilot Studio’s user-friendly interface. This AI tool helps you build applications quickly, even if you lack deep coding skills. Many users report faster app development by using natural language commands and smart features. Institutions like La Trobe University improved their productivity by integrating AI with Power Apps. You will find that Copilot Studio simplifies complex tasks, allowing you to focus on what matters most—creating efficient solutions without hassle.
Tips for success:
- Use natural language to guide the AI
- Take advantage of built-in intelligence
- Embrace automation to save time
Key Takeaways
- Microsoft Copilot Studio simplifies app development, allowing users to create applications without deep coding skills.
- Utilize natural language commands to guide the AI, making the app-building process more intuitive and efficient.
- Integrate AI capabilities to automate workflows and enhance productivity within your organization.
- Set up the necessary accounts and permissions in Azure to ensure a smooth integration with Copilot Studio.
- Create intelligent agents that can automate tasks and improve user interactions within your Power Apps.
- Monitor agent performance regularly to identify issues and optimize functionality for better user experiences.
- Leverage external data sources through APIs to enrich your app's capabilities and provide real-time information.
- Embrace automation and AI-driven features to streamline processes, allowing you to focus on higher-value tasks.
Copilot Studio Overview
What is Microsoft Copilot Studio?
Microsoft Copilot Studio is a powerful tool designed to enhance your experience with Power Apps. It allows you to create and edit copilots for applications, enabling you to interact using natural language. This functionality streamlines the app development process, making it more intuitive. You can generate app components without needing extensive coding knowledge. This democratization of technology empowers you to innovate and improve productivity within your organization.
The significance of Microsoft Copilot Studio extends beyond just ease of use. It transforms how you approach software development, especially in the context of low-code/no-code platforms. With its user-friendly interface, you can create applications quickly, even if you lack technical expertise. The integration of AI capabilities allows you to build intelligent applications that analyze data and automate workflows seamlessly.
Key Features of Copilot Studio
Microsoft Copilot Studio stands out due to its unique features that differentiate it from other automation tools. Here are some key features that enhance your experience:
| Feature | Description |
|---|---|
| AI Integration | Integrate artificial intelligence to support tasks and workflows. |
| Self-Service Solutions | Create intelligent chatbots or other tools that can answer common queries. |
| Custom Copilots | Tailor AI-driven copilots to specific business functions for improved efficiency. |
One of the most notable features is the 'Computer Use' capability. This feature allows AI agents to interact with any application that has a graphical user interface. It enhances automation by enabling agents to click buttons, select menus, and type into fields. This means you can automate tasks without needing available APIs, streamlining processes and reducing manual effort.
The 'Computer Use' feature also allows agents to handle complex dynamic interfaces, overcoming traditional limitations of robotic process automation. This capability makes automation accessible to users without programming skills, allowing you to focus on more strategic tasks.
Prerequisites to Control My Power App
Before you can effectively control your Power App with Microsoft Copilot Studio, you need to ensure that you have the right accounts and permissions set up. This preparation is crucial for a smooth integration and operation of the AI capabilities.
Required Accounts and Permissions
To get started, follow these steps to set up the necessary accounts and permissions:
- Create an app registration in the Azure portal.
- Configure API permissions in the Azure portal:
- Add a permission for the Power Platform API.
- Select Delegated permissions and choose
CopilotStudio.Copilots.Invoke. - Grant admin consent for your organization.
- Add the redirect URL and configure token settings in the Azure portal.
These steps ensure that your application can communicate securely with Copilot Studio, allowing you to control your Power App effectively.
Setting Up Your Power Apps Environment
Setting up your Power Apps environment requires careful attention to security protocols. Here are some key considerations:
- Implement data loss prevention features to manage data security.
- Establish governance controls to manage the use of generative AI features.
- Disable agent publishing to restrict the ability to publish agents using generative AI.
- Restrict data movement across geographic locations for Copilot Studio features.
- Use the Microsoft 365 admin center to govern AI actions and agents.
- Utilize Customer Lockbox for secure access to customer data.
Additionally, consider implementing web content filtering to control access based on categories and URLs. You should also apply threat intelligence filtering to block access to known malicious sites. These measures help protect your organization while using Copilot Studio.
When preparing your environment, establish an application lifecycle management (ALM) strategy. This strategy should include at least three environments: development, test, and production. Configure the production environment as a production type environment and others as sandbox type environments. Secure every environment by applying an Entra Security Group to limit access. Following ALM golden rules ensures a consistent process, such as not customizing outside of a development environment and using environment variables for settings.
By taking these steps, you will create a secure and efficient environment to control your Power App with Copilot Studio.
Connecting Power Apps with Copilot Studio

Step-by-Step Connection Process
Connecting your power apps with Microsoft Copilot Studio involves several clear steps. Follow this process carefully to ensure a smooth setup:
- Sign in to Power Apps using your organizational account.
- Select the environment where you want to install the Copilot Studio Kit. This step ensures you work in the correct workspace.
- Navigate to Solutions in the left-hand menu.
- Choose Import Solution to bring the Copilot Studio Kit into your environment.
- Click Browse and select the CopilotStudioKit_managed.zip* file you downloaded.
- Click Next and review the solution information.
- If prompted, create the required connections. These connections allow your power apps to communicate with Copilot Studio.
- Click Import and wait for the operation to complete.
- After importing, go back to Solutions and select Default Solution.
- Open Connection references and locate Copilot Studio Kit - Dataverse.
- Edit this connection reference to ensure all required fields are filled.
- If needed, create a new connection to Dataverse by selecting the appropriate option.
- Save your changes.
- Optionally, follow further configuration steps to customize your setup based on your organization's needs.
Once you complete these steps, your power apps will link with Copilot Studio, enabling you to control and automate tasks more effectively. This connection also allows you to leverage power automate flows within your apps, enhancing automation capabilities.
Verifying the Connection
After connecting power apps with Copilot Studio, you need to verify that the connection works correctly. Use these methods to confirm your setup:
- Run the command
pac connection listin your command line interface. This command lists all your current connections. Look for a connection with the API ID/providers/Microsoft.PowerApps/apis/shared_microsoftcopilotstudio. Its presence means your connection exists. - If you do not find this connection, create a new one through the Power Apps maker portal. Follow the instructions to add the Copilot Studio connection manually.
- Open your agent in Copilot Studio and publish it. Then, navigate to Channels and select Web app. Check the connection string and agent name to confirm they match your power apps environment.
- Test your power automate flows triggered by the agent to ensure they run as expected.
If you encounter issues during verification, consider these troubleshooting tips:
Troubleshooting Tips:
- Confirm your Azure portal app registration has the correct Web Redirect URI.
- Verify that the supported account types allow both organizational and personal accounts.
- Check that the
User.Readpermission is granted in API permissions.- Ensure the client secret is correctly configured in your agent settings.
- Enable Implicit grant and Hybrid flows for token generation in Azure.
- Confirm that the custom scope is created and enabled in API settings.
- Make sure your bot or agent is published with the correct configuration.
Following these steps helps you maintain a secure and functional connection between your power apps and Copilot Studio. This setup allows you to harness the full power of AI-driven automation and control within your applications. Remember, integrating power automate flows within this environment can further streamline your workflows and boost productivity.
Note: Always keep your power automate desktop and related tools updated to ensure compatibility and security when working with Copilot Studio.
By carefully connecting and verifying your power apps with Copilot Studio, you unlock new possibilities for automation and intelligent app control.
Customizing Your Power App with Copilot Studio

Creating and Configuring Agents
When you customize your Power App with copilot, creating and configuring agents plays a vital role. Agents act as intelligent assistants that interact with users and automate tasks within your canvas app. You can design agents to handle specific workflows, answer questions, or trigger actions, making your app more dynamic and user-friendly.
Follow these best practices to create and configure effective agents:
Manage Context and Memory
Use session variables to store user information during interactions. This approach helps your agent remember previous inputs and maintain context, which improves the conversation flow and user experience.Integrate with Power Automate
Connect your agent to Power Automate flows to automate complex tasks. For example, your agent can log IT tickets, fetch reports, or update records automatically, reducing manual effort and speeding up the app-building process.Connect to External Systems
Use Microsoft Dataverse and APIs to access data from various sources. This connection allows your agent to retrieve or update information beyond the canvas app, expanding its capabilities and making your app more powerful.Define Custom Workflows
Build multi-step workflows that validate user input and trigger conditional actions. This setup ensures your agent handles different scenarios correctly and guides users through complex processes smoothly.
By following these steps, you create agents that not only respond intelligently but also perform meaningful actions. This approach enhances your app development by embedding smart automation directly into your canvas app.
💡 Tip: Start with a simple agent and gradually add complexity. Test each step to ensure your agent behaves as expected before moving on.
Using AI for Customization
Copilot Studio uses AI to transform your app-building process. It offers suggestions and automates repetitive tasks, allowing you to focus on designing better user experiences. AI-driven customization accelerates app development and improves the overall quality of your canvas app.
Here is how AI enhances your Power Apps customization:
| Feature | Benefit |
|---|---|
| Accelerated Development | Build apps in days or hours instead of weeks, speeding up your project delivery. |
| Improved Productivity | Let AI handle repetitive tasks so you can focus on higher-value work and creativity. |
| Enhanced User Experience | Create smart apps that interact naturally and predict user needs, increasing engagement. |
Copilot also personalizes your app by tailoring responses and workflows to your business context. Each model-driven app can have a dedicated AI assistant that understands the specific needs of your users. This customization boosts productivity and user satisfaction.
You can use AI suggestions in Copilot Studio by following these steps:
Explore the Copilot Studio Dashboard
Familiarize yourself with key components like the Home Page, Bot Editor, Topics Section, Entities & Variables, Power Automate Integration, and Analytics.Create a New AI Agent
Choose a template or start fresh by naming your agent and describing its purpose.Define Core Instructions
Set the agent’s persona, response style, and business context to guide its behavior.Customize Greeting Messages
Add initial greetings and fallback responses to make interactions smooth and friendly.Build Conversational Flows
Use the drag-and-drop editor to create decision trees, define intents, and set triggers for user inputs.Manage Context and Memory
Implement session variables to keep conversations relevant and personalized.Add Actions & Automations
Integrate Power Automate flows to trigger workflows based on user requests.
By leveraging these AI-driven features, you simplify the app-building process and create a canvas app that feels intelligent and responsive. This approach reduces development time and enhances the overall user experience.
🚀 Remember, AI in copilot is your partner in app development. Use its suggestions to refine your app continuously and deliver solutions that truly meet your users’ needs.
Invoking Agents in Power Apps
Triggering Agents
You can trigger agents in Power Apps using both manual and automatic methods. Understanding these methods helps you utilize your copilot studio agent effectively. Here are some recommended approaches to trigger agents:
- Define triggers and payload instructions carefully to avoid conflicts.
- Test all changes involving event triggers to ensure proper functionality.
- Use agent instructions to guide how the agent should respond to triggers.
- Modify existing payload instructions using Power Automate as needed.
- Always test triggers before publishing to confirm expected behavior.
By following these steps, you ensure that your agents respond accurately to user inputs and events. This process enhances the overall user experience and makes your applications more interactive.
Monitoring Agent Actions
Monitoring the actions of your copilot studio agent is crucial for maintaining reliability and performance. You can implement several best practices to ensure your agents operate smoothly:
- Cost Optimization: Assign appropriate models to agents based on task complexity. Monitor token consumption and apply context compaction to reduce costs.
- Observability and Testing: Instrument agent operations and track performance metrics. Design testable interfaces to facilitate ongoing evaluation.
- Reliability: Implement timeout and retry mechanisms. Validate agent output and consider circuit breaker patterns to handle failures gracefully.
- Security: Use authentication and secure networking. Follow the principle of least privilege to protect sensitive data.
By actively monitoring your agents, you can identify issues early and make necessary adjustments. This proactive approach ensures that your applications remain efficient and responsive.
Troubleshooting Common Issues
Connection Failures
Connection failures can disrupt your experience with Copilot Studio and Power Apps. Here are some common issues you might encounter:
- Inconsistent responses from the Copilot Studio Agent: You may notice that the agent behaves differently in Microsoft Teams compared to Copilot Studio.
- User-specific connection problems: Some app users can interact with the agent while others cannot, even with identical configurations.
- Connection management anomalies: You might see different connection statuses, and non-admin users may access connection management settings.
To resolve these issues, check user permissions, manage connections properly, and ensure the agent is published correctly. Following these steps can help you maintain a stable connection and enhance your overall experience.
Agent Misbehavior
Agent misbehavior can hinder the functionality of your Power App. Here are some common causes of this issue:
- Infinite loops: The agent may repeat the same actions endlessly.
- Incorrect task decomposition: The agent might break down complex tasks into too many subtasks.
- Context loss: The AI agent could forget results from previous steps.
- Inefficient tool usage: The agent may call APIs with incorrect parameters.
- Poor error handling: The agent might stop or ignore errors.
- Inaccurate result interpretation: The agent could misunderstand responses from tools.
- Excessive plan detail: The agent may create overly detailed plans for simple tasks.
- Wrong priorities: The agent might focus on unimportant details.
- Lack of final result validation: The agent may not verify task completion.
- Hallucinations: The agent could invent results of actions that weren't performed.
To address these issues, follow these steps:
- Conduct thorough testing of the agent.
- Analyze agent actions to identify error patterns.
- Create test cases for each type of bug.
- Test edge cases and failure scenarios.
- Ensure agents have built-in correction or validation tools.
Using a four-criteria framework can help you evaluate whether to use an agent or a workflow. This framework considers task ambiguity, cost-value balance, minimum capability tests, and error management strategies.
Performance Optimization Techniques
To optimize performance and troubleshoot issues effectively, consider these techniques:
- Use the App Checker to identify syntax errors and issues in formulas.
- Utilize the Power Apps Monitor for real-time logging and tracing to troubleshoot issues.
- Enable Formula-Level Error Management to handle and log errors effectively.
- Write formulas that can be delegated to the cloud datasource for faster data operations.
- Reduce unnecessary data calls and limit controls to enhance performance.
By implementing these strategies, you can improve the reliability and efficiency of your Power App, ensuring a smoother experience with Copilot Studio.
Advanced Control Techniques
Integrating External Data
Integrating external data sources into your Power Apps can significantly enhance functionality. You can use REST APIs to connect various services, allowing your app to access real-time data. For example, connecting to a weather API can provide live updates directly within your application. This integration showcases the potential of external data sources to enrich your app-building experience.
Here are some effective strategies for integrating external data sources:
- Leverage Microsoft Dataverse for its strong delegation and integration capabilities.
- Use REST API integration to connect with external systems.
- Engage Power Apps consultants for optimization and alignment with business goals.
When integrating external data, consider the following table that outlines different data sources and their characteristics:
| Data Source | Delegation | Scalability | Best For |
|---|---|---|---|
| Microsoft Dataverse | Excellent | Enterprise-grade | Enterprise apps, complex data models, Power Automate integration |
| SQL Server / Azure SQL | Good | High | Large datasets, complex queries, existing SQL infrastructure |
| Custom APIs / REST | None (app-side) | Varies | Bespoke integrations with external systems |
Security is a crucial aspect when integrating external data. The data remains confined within your app and does not transmit to Microsoft Graph. This approach safeguards sensitive information, ensuring that external data privacy is maintained. Copilot Studio supports a 'bring your own protection' model, allowing you to integrate your preferred security solutions tailored to your needs.
Automating Workflows
Automating complex workflows in Power Apps with Copilot Studio can streamline processes and enhance productivity. Here are recommended steps to achieve effective automation:
- Manage Context and Memory: Use session variables to store user information and ensure contextual awareness for personalized experiences.
- Integrate Power Automate: Trigger workflows through Power Automate for actions like submitting IT tickets or pulling reports.
- Connect to External Systems: Utilize APIs or Dataverse to integrate with external systems for data retrieval and action triggering.
- Define Custom Workflows: Create multi-step workflows that include validation, conditional logic, and approval processes.
- Enhance Functionality with Integrations: Extend capabilities by integrating with Microsoft 365 applications and third-party platforms.
Real-world examples of workflow automation include:
- A field service app that automates responses to common repair questions and submits replacement requests.
- A customer onboarding portal that guides new users through setup steps and pulls their account information.
- An HR helpdesk app that automates responses to internal policy questions and initiates leave requests.
When deploying automations, emphasize governance, security, and operational best practices. This focus ensures a smooth and secure implementation, allowing you to maximize the benefits of the Microsoft Power Platform.
Incorporating Microsoft Copilot Studio into your Power Apps development can transform your approach to app control. You gain significant advantages over traditional methods, such as improved efficiency and enhanced integration capabilities.
- Real-world language in virtual agents boosts user satisfaction.
- Custom analytics provide insights into performance, aiding decision-making.
- Automated agents reduce basic queries, allowing human agents to focus on complex issues.
By leveraging these features, you can create responsive applications that meet user needs effectively. Embrace this innovative tool to elevate your app development experience.
FAQ
What is Microsoft Copilot Studio?
Microsoft Copilot Studio is an AI tool that enhances your Power Apps experience. It allows you to create and manage intelligent agents that automate tasks and improve app functionality.
How do I set up Copilot Studio?
To set up Copilot Studio, create an app registration in Azure, configure API permissions, and import the Copilot Studio Kit into your Power Apps environment.
Can I use Copilot Studio without coding skills?
Yes! Copilot Studio is designed for users without extensive coding knowledge. You can use natural language commands to build and customize your applications.
What types of tasks can Copilot Studio automate?
Copilot Studio can automate various tasks, such as data entry, form submissions, and interactions with legacy applications. It enhances efficiency by reducing manual effort.
How do I troubleshoot connection issues?
To troubleshoot connection issues, check user permissions, verify API settings, and ensure your agent is published correctly. Use the command pac connection list to confirm connections.
Is my data secure with Copilot Studio?
Yes, Copilot Studio prioritizes data security. Implement governance controls, data loss prevention features, and secure access protocols to protect sensitive information.
Can I integrate external data sources?
Absolutely! You can integrate external data sources using REST APIs or Microsoft Dataverse. This integration enhances your app's functionality and provides real-time data access.
How can I monitor agent performance?
You can monitor agent performance by tracking metrics, analyzing logs, and implementing testing protocols. Regularly review agent actions to ensure optimal functionality.
🚀 Want to be part of m365.fm?
Then stop just listening… and start showing up.
👉 Connect with me on LinkedIn and let’s make something happen:
- 🎙️ Be a podcast guest and share your story
- 🎧 Host your own episode (yes, seriously)
- 💡 Pitch topics the community actually wants to hear
- 🌍 Build your personal brand in the Microsoft 365 space
This isn’t just a podcast — it’s a platform for people who take action.
🔥 Most people wait. The best ones don’t.
👉 Connect with me on LinkedIn and send me a message:
"I want in"
Let’s build something awesome 👊
Copilot is no longer just that chatty sidekick hallucinating email drafts. With Copilot Studio’s new “Computer Use” feature, Microsoft has effectively given your AI hands and eyes. It can now physically operate a Windows machine – clicking, typing, dragging, launching your Power Apps and submitting real forms like a disturbingly obedient digital intern.
In this episode, we dive deep into how Computer Use turns Copilot from a commentator into an operator, bridging the gap between cloud automation and messy, legacy UIs that never got an API. We walk through setup, watch the AI hilariously struggle with date pickers, then get serious about governance, Entra ID identities, Purview, Defender, and Zero Trust.
By the end, you’ll know how to let an AI agent drive your Power Apps without handing it the corporate nuclear codes.
Key Topics We Cover
What “Computer Use” actually is in Copilot Studio
How Copilot uses synthetic mouse & keyboard input
Real-time computer vision + reasoning over your Windows desktop
Why this is different from Power Automate flows and classic connectors
Why this is a big deal for Microsoft 365 & Power Apps
Automating legacy desktop apps and unintegrated intranet portals
Turning “no API” systems into AI-controllable systems
Moving from AI that talks about work to AI that does the work
Step-by-Step Setup: Enabling Computer Use Safely
Windows 10/11 Pro+ and Power Automate Desktop Machine Runtime
Registering your machine in Power Automate → Monitor → Machines
Enabling “Computer Use” and version requirements
Local vs domain accounts and the security implications
Live Behavior: Watching the AI Struggle (and Learn)
The AI opening your Power App like a slightly drunk intern
Misclicks, retries, and the eternal battle with date pickers
Why every run is different: agentic reasoning vs rigid scripts
Seeing adaptation in real time: “Fine, I’ll just type the date.”
The Governance Catch: When Agents Get Permissions
Copilot agents as real Entra ID identities (not shadow tokens)
How they inherit your license, your routes, your access
Using Microsoft Purview, Defender, and Entra ID to:
Limit data exposure
Detect weird behavior
Shut down rogue runs in real time
Why audit logs, traceability, and Zero Trust boundaries are non-negotiable
Building Responsible Agentic Workflows
Sandbox first: green-zone machines, cloned Power Apps, no real data
Least privilege: narrow Entra agent IDs, segmented environments
Human-in-the-loop approvals for critical actions
Treating agents as synthetic employees with onboarding, scope, and offboarding
Turning “governance” from buzzword into operational habit
What This Means for the Future of Work
AI as a digital workforce of narrow specialists
Copilot moving from advisor → operator → orchestrator
Why “autonomy requires discipline” is the core design principle
Who This Episode Is For
Power Apps & Power Automate builders curious about Copilot Studio
Microsoft 365 architects designing secure AI automation patterns
Security & compliance teams worried about AI clicking the wrong thing
IT leaders evaluating agentic AI and AI-driven RPA in their org
Anyone who has ever thought, “If only someone could just click through this app for me…”

Founder of m365.fm, m365.show and m365con.net
Mirko Peters is a Microsoft 365 expert, content creator, and founder of m365.fm, a platform dedicated to sharing practical insights on modern workplace technologies. His work focuses on Microsoft 365 governance, security, collaboration, and real-world implementation strategies.
Through his podcast and written content, Mirko provides hands-on guidance for IT professionals, architects, and business leaders navigating the complexities of Microsoft 365. He is known for translating complex topics into clear, actionable advice, often highlighting common mistakes and overlooked risks in real-world environments.
With a strong emphasis on community contribution and knowledge sharing, Mirko is actively building a platform that connects experts, shares experiences, and helps organizations get the most out of their Microsoft 365 investments.








