Category: Power Automate

copilot-studio, Power Automate

Computer-Using Agents (CUAs) in Microsoft Copilot Studio

Goloknath

Computer-Using Agents (CUAs) in Microsoft Copilot Studio

These are agentic AI systems designed to “see, understand, and act” across web and desktop applications, specifically for complex UI automation where traditional APIs do not exist.

Real-time User Journey

The user journey for a CUA shifts from writing rigid scripts to delegating natural language instructions:

  1. Instruction: A user tells the agent, “Every night at 11 PM, log into the vendor portal, download the invoice, and enter the data into our desktop ERP system.”
  2. Autonomous Authentication: The agent retrieves encrypted logins from Azure Key Vault and signs in to both the website and the legacy desktop app without human intervention.
  3. Adaptive Action: The agent “sees” the screen. Even if the vendor website has updated its layout or a new pop-up appears, the agent uses its reasoning model (e.g., Claude 3.5 Sonnet or OpenAI) to navigate the change.
  4. Cloud Execution: The task runs on a managed Cloud PC pool (Windows 365), meaning the user’s local machine isn’t tied up.
  5. Audit & Review: The user checks the Session Replay the next morning to see a step-by-step video/screenshot log of exactly what the agent clicked and why.

Step-by-Step: How to Enable

To set up a computer-using agent in a US-based Copilot Studio environment:

  • Step 1: Create the Agent: Open Microsoft Copilot Studio and create a new agent or open an existing one.
  • Step 2: Add the Computer Use Tool: Navigate to Tools > Add tool > New tool and select Computer Use.
  • Step 3: Define the Task: Write a natural language description of the workflow the agent should perform.
  • Step 4: Configure Intelligence & Security: * Select your model (e.g., Anthropic Claude Sonnet 4.5 for dynamic UIs or OpenAI for multi-step web flows).
    • Set up Built-in Credentials (linked to Azure Key Vault) for secure, unattended logins.
  • Step 5: Provision Infrastructure: Set up a Cloud PC pool (managed Windows 365 for Agents) to handle the execution at scale.
  • Step 6: Publish: Deploy the agent for autonomous or attended runs.

Infographic: The CUA Ecosystem

This infographic summarizes the key components that allow CUAs to automate UI at scale:

Visual & LogicSecurity & AccessScale & Monitoring
Model ChoiceBuilt-in CredentialsCloud PC Pools
Uses Claude 4.5 or OpenAI to interpret screens & dynamic dashboards.Encrypted logins via Azure Key Vault for unattended runs.Managed Windows 365 machines that scale with demand.
Solves: Brittle UI changesSolves: Auth bottlenecksSolves: Hardware overhead

References

Computer-using agents now deliver more secure UI automation at scale
AI Foundry, contact-centre, copilot-studio, Customer Experience, Customer-service, Power Apps, Power Automate

Microsoft Ignite 2025: Transitioning to the Era of Agentic Business Transformation

Goloknath

Microsoft Ignite 2025: Transitioning to the Era of Agentic Business Transformation

The central theme for November 2025 was the evolution of Copilot Studio into a foundation for “agentic” workflows—where AI agents don’t just assist but autonomously drive business processes with human oversight.

Real-time User Journey: Human-in-the-Loop (HITL)

One of the most significant “real-time” journeys introduced is the Human-in-the-Loop capability:

  1. Autonomous Start: An agent begins a multi-step workflow (e.g., processing a high-value procurement order).
  2. Pause for Judgment: The agent reaches a step requiring approval or specific context (e.g., the order exceeds a price threshold).
  3. Real-time Request: The agent automatically pauses and sends a structured Outlook form to a designated human reviewer.
  4. Human Action: The person reviews the details and submits their input/approval via the form.
  5. Resumption: The agent receives the response and immediately resumes the workflow using the human’s input as a parameter to complete the task.

Step-by-Step: How to Enable GPT-5 Chat

GPT-5 Chat became generally available in November 2025 for US and EU customers. To enable it:

  • Step 1: Open your agent’s Overview page in Microsoft Copilot Studio.
  • Step 2: Locate the Model Selection settings.
  • Step 3: Select GPT-5 Chat from the dropdown menu to set it as the primary model.
  • Step 4 (Experimental): For US customers wanting to test the latest reasoning, you can toggle on GPT-5.2 (Experimental) in early release environments for improved code generation and multilingual support.

Infographic: The New Agentic Capabilities

The November 2025 update expanded the platform across three core pillars:

Intelligence (Models)Action (Skills)Governance (Control)
GPT-5 GA: Higher accuracy and instruction-following.Human-in-the-Loop: Agents pause for human approval.Microsoft Agent 365: The unified control plane for agents.
Model Choice: Multi-model support (OpenAI, Anthropic, xAI).Action Groups: One-click sets for Outlook/SharePoint tools.Entra Agent ID: Identity-based security for every agent.
MCP Integration: Connects to 1,400+ external systems.Producer Skills: Generate Word, Excel, and PPT files.Defender for Agents: Real-time threat protection.

References

What’s new in Microsoft Copilot Studio: November 2025