Chris Gillum
c67d3523ae
* Filter empty AIContent from durable agent state responses Prevent opaque AIContent objects (e.g., with only RawRepresentation set) from being stored in durable entity state, where they serialize to empty JSON payloads. Base AIContent instances are kept only if they have Annotations or AdditionalProperties. Fixes https://github.com/microsoft/agent-framework/issues/4481 * Update CHANGELOG.md and fix linter violation
Microsoft.Agents.AI.DurableTask
The Microsoft Agent Framework provides a programming model for building agents and agent workflows in .NET. This package, the Durable Task extension for the Agent Framework, extends the Agent Framework programming model with the following capabilities:
- Stateful, durable execution of agents in distributed environments
- Automatic conversation history management
- Long-running agent workflows as "durable orchestrator" functions
- Tools and dashboards for managing and monitoring agents and agent workflows
These capabilities are implemented using foundational technologies from the Durable Task technology stack:
- Durable Entities for stateful, durable execution of agents
- Durable Orchestrations for long-running agent workflows
- The Durable Task Scheduler for managing durable task execution and observability at scale
This package can be used by itself or in conjunction with the Microsoft.Agents.AI.Hosting.AzureFunctions package, which provides additional features via Azure Functions integration.
Install the package
From the command-line:
dotnet add package Microsoft.Agents.AI.DurableTask
Or directly in your project file:
<ItemGroup>
<PackageReference Include="Microsoft.Agents.AI.DurableTask" Version="[CURRENTVERSION]" />
</ItemGroup>
You can alternatively just reference the Microsoft.Agents.AI.Hosting.AzureFunctions package if you're hosting your agents and orchestrations in the Azure Functions .NET Isolated worker.
Usage Examples
For a comprehensive tour of all the functionality, concepts, and APIs, check out the Azure Functions samples.
Feedback & Contributing
We welcome feedback and contributions in our GitHub repo.