agent-framework/dotnet/samples/GettingStarted/FoundryAgents/FoundryAgents_Step06_PersistedConversations/README.md

Persisted Conversations with AI Agents

This sample demonstrates how to serialize and persist agent conversation threads to storage, allowing conversations to be resumed later.

What this sample demonstrates

• Serializing agent threads to JSON
• Persisting thread state to disk
• Loading and deserializing thread state from storage
• Resuming conversations with persisted threads
• Managing agent lifecycle (creation and deletion)

Prerequisites

Before you begin, ensure you have the following prerequisites:

• .NET 10 SDK or later
• Azure Foundry service endpoint and deployment configured
• Azure CLI installed and authenticated (for Azure credential authentication)

Note: This demo uses Azure CLI credentials for authentication. Make sure you're logged in with az login and have access to the Azure Foundry resource. For more information, see the Azure CLI documentation.

Set the following environment variables:

$env:AZURE_FOUNDRY_PROJECT_ENDPOINT="https://your-foundry-service.services.ai.azure.com/api/projects/your-foundry-project" # Replace with your Azure Foundry resource endpoint
$env:AZURE_FOUNDRY_PROJECT_DEPLOYMENT_NAME="gpt-4o-mini"  # Optional, defaults to gpt-4o-mini

Run the sample

Navigate to the FoundryAgents sample directory and run:

cd dotnet/samples/GettingStarted/FoundryAgents
dotnet run --project .\FoundryAgents_Step06_PersistedConversations

Expected behavior

The sample will:

1. Create an agent named "JokerAgent" with instructions to tell jokes
2. Create a thread and run the agent with an initial prompt
3. Serialize the thread state to JSON
4. Save the serialized thread to a temporary file
5. Load the thread from the file and deserialize it
6. Resume the conversation with the same thread using a follow-up prompt
7. Clean up resources by deleting the agent