# How to separate reasoning(chain of thoughts) and final response in LangChain v1 Agent (create\_agent)?

**URL:** <https://forum.langchain.com/t/how-to-separate-reasoning-chain-of-thoughts-and-final-response-in-langchain-v1-agent-create-agent/2252>\
**Category:** LangChain Academy\
**Tags:** python-help\
**Created:** [November 18, 2025, 11:30pm UTC](https://forum.langchain.com/t/how-to-separate-reasoning-chain-of-thoughts-and-final-response-in-langchain-v1-agent-create-agent/2252 "2025-11-18T23:30:04Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![UK59](https://avatars.discourse-cdn.com/v4/letter/u/edb3f5/32.png) [@UK59](https://forum.langchain.com/u/UK59)\
**Post date:** [November 18, 2025, 11:30pm UTC](https://forum.langchain.com/t/how-to-separate-reasoning-chain-of-thoughts-and-final-response-in-langchain-v1-agent-create-agent/2252/1 "2025-11-18T23:30:04Z")

</div>

Hi everyone,

I’m using **LangChain v1** and creating an agent using `create_agent` with a set of tools.  
The agent works as expected — it reasons through the problem, makes tool calls, and eventually generates an answer.

However, I’m facing an issue:  
➡ The agent doesn’t clearly **bifurcate between its reasoning steps(by reasoning I mean the chain of thoughts) and the final response**.

I’m using an async stream handler that invokes the agent like this:

```auto
async for token, metadata in agent.astream(
    {"messages": [{"role": "user", "content": request.query}]},
    stream_mode="messages",
):
    print(f"node: {metadata['langgraph_node']}")
    print(f"content: {token.content_blocks}")

```

I want to know if there’s a **way to distinguish or separate** :

- The agent’s _thought process / reasoning trace_, and
- The _final response_ returned to the user.

Basically, how can I capture or output both — one for internal logs and another for user-facing response?

Has anyone implemented something similar or found a recommended pattern for this in LangChain v1?  
Any insights, code snippets, or best practices would be greatly appreciated! 🙏

---

<div class="post-metadata">

**Author:** ![mcavdar](https://yyz1.discourse-cdn.com/flex007/user_avatar/forum.langchain.com/mcavdar/32/720_2.png) [@mcavdar](https://forum.langchain.com/u/mcavdar)\
**Post date:** [November 19, 2025, 1:43pm UTC](https://forum.langchain.com/t/how-to-separate-reasoning-chain-of-thoughts-and-final-response-in-langchain-v1-agent-create-agent/2252/2 "2025-11-19T13:43:18Z")

</div>

Hello @UK59  
Have you tried the example provided at [Models - Docs by LangChain](https://docs.langchain.com/oss/javascript/langchain/models#reasoning)?

---

<div class="post-metadata">

**Author:** ![UK59](https://avatars.discourse-cdn.com/v4/letter/u/edb3f5/32.png) [@UK59](https://forum.langchain.com/u/UK59)\
**Post date:** [November 19, 2025, 3:41pm UTC](https://forum.langchain.com/t/how-to-separate-reasoning-chain-of-thoughts-and-final-response-in-langchain-v1-agent-create-agent/2252/3 "2025-11-19T15:41:51Z")

</div>

Hey @mcavdar!  
Sorry for the confusion, so I’m not kinda looking for reasoning, but what I meant when I say reasoning is actually the chain of thoughts the AI does before giving back the final answer. Like how it thinks, and how it interpret the tool response and so on…

---

<div class="post-metadata">

**Author:** ![timofeyantonenko](https://yyz1.discourse-cdn.com/flex007/user_avatar/forum.langchain.com/timofeyantonenko/32/1877_2.png) [@timofeyantonenko](https://forum.langchain.com/u/timofeyantonenko)\
**Post date:** [December 8, 2025, 4:09pm UTC](https://forum.langchain.com/t/how-to-separate-reasoning-chain-of-thoughts-and-final-response-in-langchain-v1-agent-create-agent/2252/4 "2025-12-08T16:09:11Z")

</div>

Yes, having the same issue.  
I checked docs:

1. [Models - Docs by LangChain](https://docs.langchain.com/oss/python/langchain/models)
2. [Streaming API - Docs by LangChain](https://docs.langchain.com/langsmith/streaming)  
and it’s not clear how to get intermediate reasoning events.