Issue with registering nested chat functionality

[UNaeem95]

Hi everyone

i recently started on autogen i don't have much prior python experience.
the issue I'm currently having is when im trying to trigger the critique agent but it doesn't show in output.

for example flow should be

Writer -> critique -> writer -> editor

current output: writer -> editor

below is my code

writing_assistant = autogen.AssistantAgent(
    name="writing_assistant",
    system_message='''You are a world class writing assistant tasked with writing an engaging 1000 words blog post. 
    You will use all the information given to you and write based on that on the topic specified.
    You will add more depth and engaging tone to the article.
    If the user provides critique, respond with a revised version of your previous attempts.
    ''',
    llm_config=llm_config,
)

critique_assistant = autogen.AssistantAgent(
    name="critique_assistant",
    system_message='''Generate critique and recommendations on the writing. Provide detailed recommendations, 
    including requests for length, depth, style, etc..''',
    llm_config={"config_list": config_list},
)

---- some work in between to get data in informationFinal variable ----

writing_editor_assistant = autogen.AssistantAgent(
    name="writing_editor_assistant",
    system_message='''You are a Content Editor responsible for reviewing, revising, and refining written material to 
    ensure it meets quality standards, including grammar, style, tone, and accuracy. You may also fact-check content, 
    ensure consistency, and align the material with the target audience and brand guidelines before publication.''',
    llm_config={"config_list": config_list},
)


def reflection_message(recipient, messages, sender, config):
    print("Reflecting...")
    return f"Reflect and provide critique on the following writing. \n\n {recipient.chat_messages_for_summary(sender)[-1]['content']}"


nested_chat_queue = [
    {
        "recipient": critique_assistant,
        "message": reflection_message,
        "max_turns": 1,
    },
]

user_proxy.register_nested_chats(
    nested_chat_queue,
    trigger=writing_assistant,
)

user_proxy.initiate_chats(
    [
        {
            "recipient": writing_assistant,
            "message": f"Write a 1000 word blog about {task} from a Company MD perspective. Use highlights,"
                       f"bullets, and sections.Elaborate on each section, and add to the given text."
                       f"Use all key points from the given information."
                       f"The information is from 3 separate sources separated by--Information from URL."
                       f" Combine all the information and present the best blog possible--: {informationFinal}",
            "clear_history": True,
            "silent": False,
            "summary_method": "last_msg",
            "max_turns": 2
        },
        {
            "recipient": writing_editor_assistant,
            "message": "Review and edit the blog post provided by the writing_assistant.",
            "clear_history": True,
            "silent": False,
            "summary_method": "last_msg",
            "max_turns": 1
        }
    ]

this is not complete code. i have done some work to get the information in variable "informationFinal" above this.
i have done something very similar in other examples and the flow is working fine there

Thanks in advance

[ekzhu]

Not sure exactly what happened as I haven't run this code. But my suggestion is to not use nested chat for this and just use sequential chats of:

1. critique_assistant <-> writing_assistant, max turns = 2
2. user_proxy <-> writing_editor_assisatnt, max turns = 1

Alternatively, you can also avoid using sequential chats all together and just write above two initiate chats directly.