Autonomous Agent Copilot Studio: How to Build One Step-by-Step

Repetitive tasks can be productivity killers. Imagine if AI could handle these tasks for you automatically and with zero intervention. With Microsoft Copilot Studio, you can now build an autonomous agent that works without your involvement. In this tutorial, we’ll show you exactly how to create an autonomous agent in Copilot Studio that captures Microsoft Forms responses and instantly creates tasks in Planner.

Use Case Overview: Automate Class Scheduling from Form Responses

Our goal is to automate task creation based on form submissions. Here’s how:

1. User fills out a Microsoft Form to choose a class and a preferred date.
2. Form data is stored in an Excel file on OneDrive.
3. Copilot Studio agent detects a new response and pulls this data.
4. The agent creates a task in Microsoft Planner with class name and date.
5. Team members get notified, and no manual work is needed.

Step-by-Step: Building Your Autonomous Agent

1. Create a Microsoft Form

Start by designing a simple form with two questions:

• What Microsoft tool do you want to learn?
• When would you like to attend the class?

Responses are stored in an Excel file accessible via OneDrive.

Tip: Go to “View Responses” in Microsoft Forms and download the linked Excel workbook. Copy the sharing link—you’ll need it later.

2. Set Up Your Agent in Copilot Studio

1. Open Microsoft Copilot Studio and create a new agent.
2. Name it something relevant, like Class Assignment Agent.
3. Use this description:
   “Assigns class tasks to Planner and notifies the team when a form is completed.”

 Enable the following in the settings:

• Generative Orchestration
• Deep Reasoning
• (Optional) Multi-Agent Connections for future scalability

3. Define the Trigger

• Trigger: “When a Microsoft Form is submitted”
• Select your form (e.g., Pick Your Class)
• Rename the trigger for clarity: When Class Picker Form is Completed

Note: Ensure you’re signed into both Forms and Copilot Studio. You might need to re-authenticate occasionally.

4. Add Tools (Previously Called Actions)

Tool 1: Pull Form Data from Excel

• Tool: List rows present in a table (Excel Online for Business)
• Set the location:
  • File Location: OneDrive
  • File Name: Pick Your Class.xlsx
  • Table: FormsTable or your Power Query table
• Name the tool: List Rows in Pick Your Class Table

Tool 2: Get the Most Recent Row

• Tool: Get a row
• Use ID as the key column (change it to numeric in Power Query if needed)
• Name: Get a Row - Pick Your Class

5. Integrate with Microsoft Planner

Tool 3: List Plans for a Group

• Tool: List plans for a group
• Use your Microsoft 365 Group (e.g., Learn with the Nerds)
• Name: List Plans - Class Picker

Tool 4: Create a Task in Planner

• Tool: Create a task
• Group ID: Your Microsoft 365 group
• Plan ID: Target Planner plan
• Title: Dynamically fill with AI (class name)
• Additional Input: Start Date Time = preferred class date
• Name: Create Task - Class Picker

6. Configure the Agent Instructions

Go to Overview → Edit Instructions, and write:

When Class Picker Form is completed is triggered:
1. Run /List Rows in Pick Your Class Table
2. Run /Get a Row - Pick Your Class to get class choice and date preference from the most recently updated row
   - Format date as MM/DD/YYYY
3. Run /List Plans - Class Picker
4. Run /Create Task - Class Picker
   - Use name, email, class choice, and date preference to fill task
   - Input class choice as task title
   - Input date preference as start date

Note: Be specific with tool names for easy referencing and debugging.

7. Test Your Agent

• Submit a new entry in your Microsoft Form.
• Go back to Copilot Studio → Triggers → Test Trigger
• Monitor the Activity Map to ensure each step runs correctly.
• Verify in Microsoft Planner that the task has been created.

Troubleshooting Tip: If dates don’t appear correctly, double-check the format and how your Excel column is configured.

Why Use Autonomous Agents in Copilot Studio?

Autonomous agents like this one:

• Save time by automating repetitive workflows
• Reduce errors from manual entry
• Improve collaboration by ensuring tasks are instantly assigned
• Scale easily with multi-agent orchestration and integration

Final Thoughts and Next Steps

Now that you’ve seen how to create an autonomous agent with Copilot Studio, imagine the other possibilities—project approvals, support ticket routing, onboarding checklists, and more.

For more advanced tutorials, check out this full end-to-end training:
Power BI & Copilot Studio Training – Data Bear