microsoft-agents-copilotstudio-client (original) (raw)

Microsoft Agents Copilot Studio Client

The Copilot Studio Client is for connecting to and interacting with agents created in Microsoft Copilot Studio. This library allows you to integrate Copilot Studio agents into your Python applications.

This client library provides a direct connection to Copilot Studio agents, bypassing traditional chat channels. It's perfect for integrating AI conversations into your applications, building custom UIs, or creating agent-to-agent communication flows.

What is this?

This library is part of the Microsoft 365 Agents SDK for Python - a comprehensive framework for building enterprise-grade conversational AI agents. The SDK enables developers to create intelligent agents that work across multiple platforms including Microsoft Teams, M365 Copilot, Copilot Studio, and web chat, with support for third-party integrations like Slack, Facebook Messenger, and Twilio.

Release Notes

Packages Overview

We offer the following PyPI packages to create conversational experiences based on Agents:

Package Name	PyPI Version	Description
microsoft-agents-activity		Types and validators implementing the Activity protocol spec.
microsoft-agents-hosting-core		Core library for Microsoft Agents hosting.
microsoft-agents-hosting-aiohttp		Configures aiohttp to run the Agent.
microsoft-agents-hosting-teams		Provides classes to host an Agent for Teams.
microsoft-agents-storage-blob		Extension to use Azure Blob as storage.
microsoft-agents-storage-cosmos		Extension to use CosmosDB as storage.
microsoft-agents-authentication-msal		MSAL-based authentication for Microsoft Agents.

Additionally we provide a Copilot Studio Client, to interact with Agents created in CopilotStudio:

Package Name	PyPI Version	Description
microsoft-agents-copilotstudio-client		Direct to Engine client to interact with Agents created in CopilotStudio

Installation

pip install microsoft-agents-copilotstudio-client

Quick Start

Basic Setup

Standard Environment-Based Connection

Code below from the main.py in the Copilot Studio Client

def create_client(): settings = ConnectionSettings( environment_id=environ.get("COPILOTSTUDIOAGENT__ENVIRONMENTID"), agent_identifier=environ.get("COPILOTSTUDIOAGENT__SCHEMANAME"), cloud=None, copilot_agent_type=None, custom_power_platform_cloud=None, ) token = acquire_token( settings, app_client_id=environ.get("COPILOTSTUDIOAGENT__AGENTAPPID"), tenant_id=environ.get("COPILOTSTUDIOAGENT__TENANTID"), ) copilot_client = CopilotClient(settings, token) return copilot_client

DirectConnect URL Mode (Simplified Setup)

For simplified setup, you can use a DirectConnect URL instead of environment-based configuration:

def create_client_direct(): settings = ConnectionSettings( environment_id="", # Not needed with DirectConnect URL agent_identifier="", # Not needed with DirectConnect URL direct_connect_url="https://api.powerplatform.com/copilotstudio/dataverse-backed/authenticated/bots/your-bot-id" ) token = acquire_token(...) copilot_client = CopilotClient(settings, token) return copilot_client

Advanced Configuration Options

settings = ConnectionSettings( environment_id="your-env-id", agent_identifier="your-agent-id", cloud=PowerPlatformCloud.PROD, copilot_agent_type=AgentType.PUBLISHED, custom_power_platform_cloud=None, direct_connect_url=None, # Optional: Direct URL to agent use_experimental_endpoint=False, # Optional: Enable experimental features enable_diagnostics=False, # Optional: Enable diagnostic logging (logs HTTP details) client_session_settings={"timeout": aiohttp.ClientTimeout(total=60)} # Optional: aiohttp settings )

Diagnostic Logging Details: When enable_diagnostics=True, the CopilotClient logs detailed HTTP communication using Python's logging module at the DEBUG level:

• Pre-request: Logs the full request URL (>>> SEND TO {url})
• Post-response: Logs all HTTP response headers in a formatted table
• Errors: Logs error messages with status codes

To see diagnostic output, configure your Python logging:

import logging logging.basicConfig(level=logging.DEBUG)

Experimental Endpoint Details: When use_experimental_endpoint=True, the CopilotClient will automatically capture and use the experimental endpoint URL from the first response:

• The server returns the experimental endpoint in the x-ms-d2e-experimental response header
• Once captured, this URL is stored in settings.direct_connect_url and used for all subsequent requests
• This feature is only active when use_experimental_endpoint=True AND direct_connect_url is not already set
• The experimental endpoint allows access to pre-release features and optimizations

Start a Conversation

Simple Start

The code below is summarized from the main.py in the Copilot Studio Client. See that sample for complete & working code.

copilot_client = create_client() async for activity in copilot_client.start_conversation(emit_start_conversation_event=True): if activity.type == ActivityTypes.message: print(f"\n{activity.text}")

Ask questions

async for reply in copilot_client.ask_question("Who are you?", conversation_id): if reply.type == ActivityTypes.message: print(f"\n{reply.text}")

Start with Advanced Options (Locale Support)

from microsoft_agents.copilotstudio.client import StartRequest

Create a start request with locale

start_request = StartRequest( emit_start_conversation_event=True, locale="en-US", # Optional: specify conversation locale conversation_id="custom-conv-id" # Optional: provide your own conversation ID )

async for activity in copilot_client.start_conversation_with_request(start_request): if activity.type == ActivityTypes.message: print(f"\n{activity.text}")

Send Activities

Send a Custom Activity

from microsoft_agents.activity import Activity

activity = Activity( type="message", text="Hello, agent!", conversation={"id": conversation_id} )

async for reply in copilot_client.send_activity(activity): print(f"Response: {reply.text}")

Execute with Explicit Conversation ID

Execute an activity with a specific conversation ID

activity = Activity(type="message", text="What's the weather?")

async for reply in copilot_client.execute(conversation_id="conv-123", activity=activity): print(f"Response: {reply.text}")

Subscribe to Conversation Events

For real-time event streaming with resumption support:

from microsoft_agents.copilotstudio.client import SubscribeEvent

Subscribe to conversation events

async for subscribe_event in copilot_client.subscribe( conversation_id="conv-123", last_received_event_id=None # Optional: resume from last event ): activity = subscribe_event.activity event_id = subscribe_event.event_id # Use for resumption

if activity.type == ActivityTypes.message:
    print(f"[{event_id}] {activity.text}")

Environment Variables

Set up your .env file with the following options:

Standard Environment-Based Configuration

Required (unless using DIRECT_CONNECT_URL)

ENVIRONMENT_ID=your-power-platform-environment-id AGENT_IDENTIFIER=your-copilot-studio-agent-id APP_CLIENT_ID=your-azure-app-client-id TENANT_ID=your-azure-tenant-id

Optional Cloud Configuration

CLOUD=PROD # Options: PROD, GOV, HIGH, DOD, MOONCAKE, DEV, TEST, etc. COPILOT_AGENT_TYPE=PUBLISHED # Options: PUBLISHED, PREBUILT CUSTOM_POWER_PLATFORM_CLOUD=https://custom.cloud.com

DirectConnect URL Configuration (Alternative)

Required for DirectConnect mode

DIRECT_CONNECT_URL=https://api.powerplatform.com/copilotstudio/dataverse-backed/authenticated/bots/your-bot-id APP_CLIENT_ID=your-azure-app-client-id TENANT_ID=your-azure-tenant-id

Optional

CLOUD=PROD # Used for token audience resolution

Advanced Options

Experimental and diagnostic features

USE_EXPERIMENTAL_ENDPOINT=false # Enable automatic experimental endpoint capture ENABLE_DIAGNOSTICS=false # Enable diagnostic logging (logs HTTP requests/responses)

Experimental Endpoint: When USE_EXPERIMENTAL_ENDPOINT=true, the client automatically captures and uses the experimental endpoint URL from the server's x-ms-d2e-experimental response header. This feature:

• Only activates when direct_connect_url is not already set
• Captures the URL from the first response and stores it for all subsequent requests
• Provides access to pre-release features and performance optimizations
• Useful for testing new capabilities before general availability

Diagnostic Logging: When ENABLE_DIAGNOSTICS=true or enable_diagnostics=True, the client will log detailed HTTP request and response information including:

• Request URLs before sending
• All response headers with their values
• Error messages for failed requests

This is useful for debugging connection issues, authentication problems, or understanding the communication flow with Copilot Studio. Diagnostic logs use Python's standard logging module at the DEBUG level.

Using Environment Variables in Code

The ConnectionSettings.populate_from_environment() helper method automatically loads these variables:

from microsoft_agents.copilotstudio.client import ConnectionSettings

Automatically loads from environment variables

settings_dict = ConnectionSettings.populate_from_environment() settings = ConnectionSettings(**settings_dict)

Features

Core Capabilities

✅ Real-time streaming - Server-sent events for live responses ✅ Multi-cloud support - Works across all Power Platform clouds (PROD, GOV, HIGH, DOD, MOONCAKE, etc.) ✅ Rich content - Support for cards, actions, and attachments ✅ Conversation management - Maintain context across interactions ✅ Custom activities - Send structured data to agents ✅ Async/await - Modern Python async support

Advanced Features

✅ DirectConnect URLs - Simplified connection with direct bot URLs ✅ Locale support - Specify conversation language with StartRequest✅ Event subscription - Subscribe to conversation events with SSE resumption ✅ Multiple connection modes - Environment-based or DirectConnect URL ✅ Token audience resolution - Automatic cloud detection from URLs ✅ User-Agent tracking - Automatic SDK version and platform headers ✅ Environment configuration - Automatic loading from environment variables ✅ Experimental endpoints - Toggle experimental API features ✅ Diagnostic logging - HTTP request/response logging for debugging and troubleshooting

API Methods

Configuration Models

Connection Modes

The client supports two connection modes:

1. Environment-Based Connection (Standard)

Uses environment ID and agent identifier to construct the connection URL:

settings = ConnectionSettings( environment_id="aaaabbbb-1111-2222-3333-ccccddddeeee", agent_identifier="cr123_myagent" )

URL Pattern: https://{env-prefix}.{env-suffix}.environment.api.powerplatform.com/copilotstudio/dataverse-backed/authenticated/bots/{agent-id}/conversations

2. DirectConnect URL Mode (Simplified)

Uses a direct URL to the agent, bypassing environment resolution:

settings = ConnectionSettings( environment_id="", agent_identifier="", direct_connect_url="https://api.powerplatform.com/copilotstudio/dataverse-backed/authenticated/bots/cr123_myagent" )

Benefits:

• Simpler configuration with single URL
• Automatic cloud detection for token audience
• Works across environments without environment ID lookup
• Useful for multi-tenant scenarios

Token Audience Resolution

The client automatically determines the correct token audience:

For environment-based connections

audience = PowerPlatformEnvironment.get_token_audience(settings)

Returns: https://api.powerplatform.com/.default

For DirectConnect URLs

audience = PowerPlatformEnvironment.get_token_audience( settings=ConnectionSettings("", "", direct_connect_url="https://api.gov.powerplatform.microsoft.us/...") )

Returns: https://api.gov.powerplatform.microsoft.us/.default

Troubleshooting

Common Issues

Authentication failed

• Verify your app is registered in Azure AD
• Check that token has the correct audience scope (use PowerPlatformEnvironment.get_token_audience())
• Ensure your app has permissions to the Power Platform environment
• For DirectConnect URLs, verify cloud setting matches the URL domain

Agent not found

• Verify the environment ID and agent identifier
• Check that the agent is published and accessible
• Confirm you're using the correct cloud setting
• For DirectConnect URLs, ensure the URL is correct and complete

Connection timeout

• Check network connectivity to Power Platform
• Verify firewall settings allow HTTPS traffic
• Try a different cloud region if available
• Check if client_session_settings timeout is appropriate

Invalid DirectConnect URL

• Ensure URL includes scheme (https://)
• Verify URL format matches expected pattern
• Check for trailing slashes (automatically normalized)
• Confirm URL points to the correct cloud environment

Requirements

• Python 3.10+ (supports 3.10, 3.11, 3.12, 3.13, 3.14)
• Valid Azure AD app registration
• Access to Microsoft Power Platform environment
• Published Copilot Studio agent

Quick Links

• 📦 All SDK Packages on PyPI
• 📖 Complete Documentation
• 💡 Python Samples Repository
• 🐛 Report Issues

Sample Applications