This gallery helps Semantic Kernel (SK) developers move to the Microsoft Agent Framework (AF) with minimal guesswork. Each script pairs SK code with its AF equivalent so you can compare primitives, tooling, and orchestration patterns side by side while you migrate production workloads.

• 01_basic_chat_completion.py — Minimal SK ChatCompletionAgent and AF Agent conversation.
• 02_chat_completion_with_tool.py — Adds a simple tool/function call in both SDKs.
• 03_chat_completion_thread_and_stream.py — Demonstrates session reuse and streaming prompts.

OpenAI Assistants parity samples were removed alongside the deprecated Python assistants surface and are no longer part of this migration gallery.

• 01_basic_responses_agent.py — Basic responses agent migration.
• 02_responses_agent_with_tool.py — Tool-augmented responses workflows.
• 03_responses_agent_structured_output.py — Structured JSON output alignment.

• 01_basic_copilot_studio_agent.py — Minimal Copilot Studio agent invocation.
• 02_copilot_studio_streaming.py — Streaming responses from Copilot Studio agents.

• sequential.py — Step-by-step SK Team → AF SequentialBuilder migration.
• concurrent_basic.py — Concurrent orchestration parity.
• group_chat.py — Group chat coordination with an LLM-backed manager in both SDKs.
• handoff.py - Handoff coordination between agents.
• magentic.py — Magentic Team orchestration vs. AF builder wiring.

• fan_out_fan_in_process.py — Fan-out/fan-in comparison between SK Process Framework and AF workflows.
• nested_process.py — Nested process orchestration vs. AF sub-workflows.

Each script is fully async and the main() routine runs both implementations back to back so you can observe their outputs in a single execution.

• Python 3.10 or later.
• Access to the necessary model endpoints (Azure OpenAI, OpenAI, Azure AI, Copilot Studio, etc.).
• Installed SDKs: semantic-kernel and the Microsoft Agent Framework (pip install semantic-kernel agent-framework), or the repo’s editable packages if you are developing locally.
• Service credentials exposed through environment variables (for example OPENAI_API_KEY, AZURE_OPENAI_ENDPOINT, AZURE_OPENAI_API_KEY, or Copilot Studio auth settings).

From the repository root:

python samples/semantic-kernel-migration/chat_completion/01_basic_chat_completion.py

Every script accepts no CLI arguments and will first call the SK implementation, followed by the AF version. Adjust the prompt or credentials inside the file as necessary before running.

Advanced comparisons are split between samantic-kernel-migration/orchestrations (Sequential, Concurrent, Magentic) and samantic-kernel-migration/processes (fan-out/fan-in, nested). You can run them directly, or isolate dependencies in a throwaway virtual environment:

cd samples/semantic-kernel-migration
uv venv --python 3.10 .venv-migration
source .venv-migration/bin/activate
uv pip install semantic-kernel agent-framework
uv run python orchestrations/sequential.py
uv run python processes/fan_out_fan_in_process.py

Swap the script path for any other workflow or process sample. Deactivate the sandbox with deactivate when you are finished.

• Keep the original SK sample open while iterating on the AF equivalent; the code is intentionally formatted so you can copy/paste across SDKs.
• Sessions/conversation state are explicit in AF. When porting SK code that relies on implicit session reuse, call agent.create_session() and pass it into each run call.
• Tools map cleanly: SK @kernel_function plugins translate to AF @tool callables. Hosted tools (code interpreter, web search, MCP) are available only in AF—introduce them once parity is achieved.
• For multi-agent orchestration, AF workflows expose checkpoints and resume capabilities that SK Process/Team abstractions do not. Use the workflow samples as a blueprint when modernizing complex agent graphs.