agent-framework/python/samples/getting_started/durabletask/README.md

Durable Task Samples

This directory contains samples for durable agent hosting using the Durable Task Scheduler. These samples demonstrate the worker-client architecture pattern, enabling distributed agent execution with persistent conversation state.

• 01_single_agent: A sample that demonstrates how to host a single conversational agent using the Durable Task Scheduler and interact with it via a client.
• 04_single_agent_orchestration_chaining: A sample that demonstrates how to chain multiple invocations of the same agent using a durable orchestration.
• 05_multi_agent_orchestration_concurrency: A sample that demonstrates how to host multiple agents and run them concurrently using a durable orchestration.

Running the Samples

These samples are designed to be run locally in a cloned repository.

Prerequisites

The following prerequisites are required to run the samples:

• Python 3.9 or later
• Azure CLI installed and authenticated (az login) or an API key for the Azure OpenAI service
• Azure OpenAI Service with a deployed model (gpt-4o-mini or better is recommended)
• Durable Task Scheduler (local emulator or Azure-hosted)
• Docker installed if running the Durable Task Scheduler emulator locally

Configuring RBAC Permissions for Azure OpenAI

These samples are configured to use the Azure OpenAI service with RBAC permissions to access the model. You'll need to configure the RBAC permissions for the Azure OpenAI service to allow the Python app to access the model.

Below is an example of how to configure the RBAC permissions for the Azure OpenAI service to allow the current user to access the model.

Bash (Linux/macOS/WSL):

az role assignment create \
  --assignee "yourname@contoso.com" \
  --role "Cognitive Services OpenAI User" \
  --scope /subscriptions/<your-subscription-id>/resourceGroups/<your-resource-group-name>/providers/Microsoft.CognitiveServices/accounts/<your-openai-resource-name>

PowerShell:

az role assignment create `
  --assignee "yourname@contoso.com" `
  --role "Cognitive Services OpenAI User" `
  --scope /subscriptions/<your-subscription-id>/resourceGroups/<your-resource-group-name>/providers/Microsoft.CognitiveServices/accounts/<your-openai-resource-name>

More information on how to configure RBAC permissions for Azure OpenAI can be found in the Azure OpenAI documentation.

Setting an API key for the Azure OpenAI service

As an alternative to configuring Azure RBAC permissions, you can set an API key for the Azure OpenAI service by setting the AZURE_OPENAI_API_KEY environment variable.

Bash (Linux/macOS/WSL):

export AZURE_OPENAI_API_KEY="your-api-key"

PowerShell:

$env:AZURE_OPENAI_API_KEY="your-api-key"

Start Durable Task Scheduler

Most samples use the Durable Task Scheduler (DTS) to support hosted agents and durable orchestrations. DTS also allows you to view the status of orchestrations and their inputs and outputs from a web UI.

To run the Durable Task Scheduler locally, you can use the following docker command:

docker run -d --name dts-emulator -p 8080:8080 -p 8082:8082 mcr.microsoft.com/dts/dts-emulator:latest

The DTS dashboard will be available at http://localhost:8082.

Environment Configuration

Each sample reads configuration from environment variables. You'll need to set the following environment variables:

export AZURE_OPENAI_ENDPOINT="https://your-resource.openai.azure.com/"
export AZURE_OPENAI_CHAT_DEPLOYMENT_NAME="your-deployment-name"

Installing Dependencies

Navigate to the sample directory and install dependencies:

cd samples/getting_started/durabletask/01_single_agent
pip install -r requirements.txt

Running the Samples

Each sample follows a worker-client architecture. Most samples provide separate worker.py and client.py files, though some include a combined sample.py for convenience.

Running with separate worker and client:

In one terminal, start the worker:

python worker.py

In another terminal, run the client:

python client.py

Running with combined sample:

python sample.py

Viewing the Sample Output

The sample output is displayed directly in the terminal where you ran the Python script. Agent responses are printed to stdout with log formatting for better readability.

You can also see the state of agents and orchestrations in the Durable Task Scheduler dashboard at http://localhost:8082.