# extract_reasoning_summary_fragments

> **Function** in `langchain_azure_ai`

📖 [View in docs](https://reference.langchain.com/python/langchain-azure-ai/agents/hosting/_converters/_utils/extract_reasoning_summary_fragments)

Return the reasoning summary text fragments in a message ``content``.

When a chat model is configured to stream reasoning summaries
(e.g. ``AzureChatOpenAI(reasoning={"summary": "auto"})``), each
:class:`~langchain_core.messages.AIMessageChunk` carries a list
``content`` that may include reasoning blocks shaped
``{"type": "reasoning", "summary": [{"type": "summary_text",
"text": "<delta>"}]}``. The summary text arrives incrementally across
chunks; an empty fragment (``""``) marks the start of a new summary
section.

:func:`extract_text` deliberately ignores these blocks (they carry no
top-level ``text`` key), so reasoning needs its own extractor.

## Signature

```python
extract_reasoning_summary_fragments(
    content: Any,
) -> list[str]
```

## Parameters

| Name | Type | Required | Description |
|------|------|----------|-------------|
| `content` | `Any` | Yes | The raw ``content`` field of a LangChain message. |

## Returns

`list[str]`

The ordered list of summary text fragments. Fragments are kept as

---

[View source on GitHub](https://github.com/langchain-ai/langchain-azure/blob/7a6a664d617264df8dcb12fcfb1bacc09f21d504/libs/azure-ai/langchain_azure_ai/agents/hosting/_converters/_utils.py#L70)