Microsoft Agent Framework: The Agent Stack Settles
Microsoft Agent Framework merges Semantic Kernel and AutoGen into one open-source agent stack for .NET and Python. Both predecessors are in maintenance mode. We summarize the timeline and the API changes, then give a migration playbook and a parity-testing approach that catches behavioral drift before cutover — as of January 2026, with GA expected within months.
Two Frameworks Were One Too Many
Since 2023 Microsoft has shipped two agent stacks. Semantic Kernel is the enterprise SDK; its .NET version reached 1.0 in December 2023 with a commitment to non-breaking changes. AutoGen came out of Microsoft Research, pioneered multi-agent orchestration, and reached v0.4 in January 2025 with an event-driven redesign. Every team building agents on Microsoft tooling had to choose: production support or research-grade orchestration patterns. The two roadmaps ran explicitly in parallel.
On October 1, 2025 Microsoft resolved the fork. Agent Framework entered public preview as one open-source SDK and runtime for .NET and Python, built by the combined Semantic Kernel and AutoGen teams. Both predecessors moved to maintenance mode the same day: bug fixes, security patches and stability updates, no new features. Anyone running agents on either stack now owns a migration. The only open variables are timing and method.
What Microsoft Agent Framework Is
Agent Framework is a set of libraries for building, orchestrating and deploying agents. On .NET the packages live under Microsoft.Agents.AI on NuGet and target .NET 8 through 10, .NET Standard 2.0 and .NET Framework 4.7.2. The core abstraction is AIAgent; model access goes through the IChatClient interface from Microsoft.Extensions.AI, so any conforming provider SDK plugs in. On Python, agent-framework on PyPI is a meta package over agent-framework-core plus per-provider packages. There is no Java version.
The framework has two layers. Agents cover single-assistant scenarios: instructions, tools, threads. Workflows cover multi-agent scenarios as typed graphs with checkpointing and built-in request-response for human-in-the-loop steps — AutoGen's group-chat patterns rebuilt with production controls. At Ignite on November 18, 2025 Microsoft added durable execution, AG-UI and ChatKit protocol support, and hosted agents in Foundry Agent Service, all in preview.
Maintenance Mode Is a Deadline
Maintenance mode is precisely defined. AutoGen is community-managed and will receive no new features; its repository says so in a warning banner. Semantic Kernel 1.x continues to receive critical fixes, and Microsoft commits to supporting it for at least one year after Agent Framework reaches general availability. That sentence contains a clock. It starts at GA, and GA has not happened yet.
What maintenance mode does not mean: no forced breaking changes and no shutdown — existing workloads keep running. What it does mean: new model capabilities, connectors and protocol integrations land only in Agent Framework. The cost of staying is not instability. It is a widening feature gap and, after the support window closes, unpatched dependencies.
What Changes in the Code
Parts of the migration are mechanical. Namespaces move to Microsoft.Agents.AI; message and content types come from Microsoft.Extensions.AI. The Kernel object disappears — dependency injection is plain .NET. Service-specific agent classes collapse into a single ChatClientAgent. The table shows the mapping we use as a checklist.
Other parts are not mechanical. Prompt filters, memory connectors and custom planners have no one-to-one equivalents. AutoGen's AssistantAgent is single-turn by default; the Agent Framework agent loops over tool calls until it can answer. Ports that ignore such semantic differences compile cleanly and behave differently.
| Before (SK / AutoGen) | Agent Framework |
|---|---|
| Microsoft.SemanticKernel namespaces | Microsoft.Agents.AI |
| Kernel as service container | standard DI plus IChatClient |
| ChatCompletionAgent / AzureAIAgent / OpenAIAssistantAgent | ChatClientAgent |
| AutoGen Team and GroupChat | typed workflow graph |
| semantic-kernel / autogen on PyPI | agent-framework |
A Migration Playbook
Our playbook has six steps. First, inventory: list every agent, tool, connector, memory store and filter, with owners. Second, pin the current Semantic Kernel or AutoGen version so the baseline stops moving. Third, cut a seam: route model access through IChatClient while still on Semantic Kernel — this shrinks the later diff. Fourth, port leaf agents before orchestration; single agents translate almost directly, group chats need a redesign as workflow graphs.
Fifth, run both stacks in parallel behind a feature flag and route a fixed share of traffic to the port. Sixth, cut over per agent, not per system, and keep the rollback path alive for one release cycle. The playbook contains no step called rewrite. If a component needs a redesign, do it after parity, not during.
Parity Testing Before Cutover
Parity testing means running the same inputs against both stacks and comparing behavior, not code. Freeze the model version and sampling parameters. Record golden transcripts on the old stack: system prompts, tool-call sequences with arguments, termination reasons, final answers, token counts. Replay them on Agent Framework. Compare structure first — did the same tools fire with the same arguments — and text second, with embedding similarity or an LLM judge and explicit thresholds. Both stacks emit OpenTelemetry, so traces are directly comparable.
Parity tests do not prove equivalence. Language-model output is not deterministic, even at temperature zero, so identical answers are the wrong success criterion. What the tests reliably catch is drift you caused: a dropped system prompt, a changed tool schema, lost memory wiring, a different tool-iteration default. That class of bug is the migration risk.
What the Preview Does Not Give You
The preview is not a frozen API. The .NET release of December 19, 2025 — 1.0.0-preview.251219.1 — still contained breaking namespace changes. Some AutoGen model clients have no equivalent yet, and AutoGen's experimental distributed runtime has no counterpart; Agent Framework composes in a single process today. Semantic Kernel's Java SDK has no successor path at all. Teams on these edges should build the parity harness now and defer the port.
Outlook From January 2026
Microsoft said in October that the preview would run for several months. Package releases have arrived roughly every two weeks since. Our expectation, written in January 2026: a release candidate within the next few weeks, general availability in the first half of 2026 — most likely spring — and therefore Semantic Kernel maintenance guaranteed into 2027.
The larger prediction: the framework question in the Microsoft ecosystem is settled, and competition moves up the stack — to hosted runtimes, workflow durability and governance. We at Blue IT Systems are running the playbook above on our own Semantic Kernel codebases and will migrate after the API freeze, not before. Build the parity harness now. It is useful regardless of when GA lands.
Sources
- Introducing Microsoft Agent Framework — Microsoft Azure Blog (October 1, 2025)
- Introducing Microsoft Agent Framework (Preview) — .NET Blog (October 1, 2025)
- Semantic Kernel and Microsoft Agent Framework — Agent Framework Blog (October 7, 2025)
- Foundry Agent Service at Ignite 2025 — Microsoft Tech Community (November 18, 2025)
- Agent Framework release dotnet-1.0.0-preview.251219.1 — GitHub (December 19, 2025)