Copilot Studio

Copilot Studio integration enables you to use Copilot Studio agents within the Agent Framework.

The following example shows how to create an agent using Copilot Studio:

using System;
using Azure.Identity;
using Microsoft.Agents.AI;
using Microsoft.Agents.AI.CopilotStudio;

// Create a Copilot Studio agent using the IChatClient pattern
// Requires: dotnet add package Microsoft.Agents.AI.CopilotStudio --prerelease
var copilotClient = new CopilotStudioChatClient(
    environmentId: "<your-environment-id>",
    agentIdentifier: "<your-agent-id>",
    credential: new AzureCliCredential());

AIAgent agent = copilotClient.AsAIAgent(
    instructions: "You are a helpful enterprise assistant.");

Console.WriteLine(await agent.RunAsync("What are our company policies on remote work?"));

Tools

Copilot Studio agents run remotely: the agent definition (topics, knowledge sources, generative actions, plugins, MCP servers) is authored in the Copilot Studio portal. The Agent Framework Copilot Studio client invokes the published agent and surfaces its responses — it does not expose Agent Framework tool types (function tools, code interpreter, file search, hosted/local MCP, etc.) at the client. To extend the agent's capabilities, configure those capabilities on the Copilot Studio agent itself.

Note

Python support for Copilot Studio agents is available through the agent-framework-copilotstudio package.

Installation

pip install agent-framework-copilotstudio --pre

Configuration

Set the following environment variables for automatic configuration:

COPILOTSTUDIOAGENT__ENVIRONMENTID="<your-environment-id>"
COPILOTSTUDIOAGENT__SCHEMANAME="<your-agent-schema-name>"
COPILOTSTUDIOAGENT__AGENTAPPID="<your-client-id>"
COPILOTSTUDIOAGENT__TENANTID="<your-tenant-id>"

Create a Copilot Studio Agent

CopilotStudioAgent reads connection settings from environment variables automatically:

import asyncio
from agent_framework.microsoft import CopilotStudioAgent

async def main():
    agent = CopilotStudioAgent()

    result = await agent.run("What are our company policies on remote work?")
    print(result)

asyncio.run(main())

Tools

CopilotStudioAgent invokes a Copilot Studio agent that runs remotely. The agent's behavior — topics, knowledge sources, generative actions, plugins, MCP servers — is configured in the Copilot Studio portal, not in your Python code. The Agent Framework client does not expose Agent Framework tool types (function tools, code interpreter, file search, hosted/local MCP, etc.) at the client. To extend the agent's capabilities, configure those capabilities on the Copilot Studio agent itself.

Streaming

async def streaming_example():
    agent = CopilotStudioAgent()

    print("Agent: ", end="", flush=True)
    async for chunk in agent.run("What is the largest city in France?", stream=True):
        if chunk.text:
            print(chunk.text, end="", flush=True)
    print()

Next steps

Custom Provider