Tim Schupp

Had the same issue once, don't remember where I found the fix but this solved a similar error for me: ```js import * as toolkitRaw from "@reduxjs/toolkit"; const { configureStore...

I see that the agent executor not accepts `return_intermediate_step = True` but I'm still unclear how to use this. When setting this you cannot use `agent.run(...)` anymore as it causes:...

Ok I've realized my confusion as this issue is talking about intermediate steps for Tools. I've moved my comments on `return_intermediate_step` usage in agents to: #3091

Also looking for something like this. Using `ConversationChain` with `ConversationSummaryBufferMemory` running `conversation.memory.json()` I get the following error: ``` TypeError: Object of type 'SharedCallbackManager' is not JSON serializable ``` What does...

Yes I saw the `RedisChatMessageHistory` but currently I prefere not to use redis. Having a general interface for serializing and loading chat history would be great.

Also curious on this one, I think it's just not possible at the moment right? I mean streaming intermediate steps doesn't make that much sense, and you can already get...

Ok on second look callback manager does implement the agent callbacks already: Using ```python def new_agent_action(*args, **kwargs): print("AGENT ACTION", args, kwargs, flush=True) with get_openai_callback() as cb: cb.on_llm_new_token = new_llm_token cb.on_agent_action...

This does also work without setting `return_intermediate_step = True` thus not requiring the workaround with saving memory context!

Think this is what the [Conversation Agent](https://python.langchain.com/en/latest/modules/agents/agents/examples/conversational_agent.html) is for. I've tried it mostly identifies when not to use a tool and then normally continues the conversation.

Experiencing a similar issue, when using `keepmenu` either from command-line or directly launched from terminal. For me it converts `|` to `>`. But when I do `xdotool type "|>"` I...