Assistant in team failed to communicate with each other.

[kayue]

I was trying to replicate https://github.com/andrewyng/translation-agent in phidatahq but found it very difficult to pass result from one assistant to another assistant.

I cannot inject result from previous step to the prompt neither.

This is a friendly framework but with its limitation

from textwrap import dedent
from phi.assistant import Assistant
from phi.llm.openai import OpenAIChat
from pydantic import BaseModel, Field
from typing import Optional

source_lang = "English"
target_lang = "Traditional Chinese"
country = "Taiwan"


translation_assistant = Assistant(
    name="Initial translatior",
    role="Get an initial translation of the source text.",
    llm=OpenAIChat(model="gpt-4o", temperature=0, api_key=OPENAI_API_KEY),
    description=f"""You are an expert linguist, specializing in translation from {source_lang} to {target_lang}.""",
    instructions=[
        f"""This is an {source_lang} to {target_lang} translation, please provide the {target_lang} translation for this text.""",
        "Do not provide any explanations or text apart from the translation.",
    ],
    tools=[Hypebeast()],
    # output_model=Translation,
    show_tool_calls=False,
    read_tool_call_history=True,
    read_chat_history=True,
    debug_mode=False,
)

suggestion_assistant = Assistant(
    name="Proofreader",
    role="Reflect on the initial translation and generate suggestions.",
    llm=OpenAIChat(model="gpt-4o", temperature=0, api_key=OPENAI_API_KEY),
    description=dedent(f"""
        You are an expert linguist, specializing in translation from {source_lang} to {target_lang}.
        You will be provided with a source text and its translation and your goal is to improve the translation.
    """),
    instructions=[
        dedent(f"""
        Your task is to carefully read a source text and a translation from {source_lang} to {target_lang}, and then give constructive criticism and helpful suggestions to improve the translation. \
        The final style and tone of the translation should match the style of {target_lang} colloquially spoken in {country}.
        """),
        dedent(f"""
        When writing suggestions, pay attention to whether there are ways to improve the translation's
        (i) accuracy (by correcting errors of addition, mistranslation, omission, or untranslated text),
        (ii) fluency (by applying {target_lang} grammar, spelling and punctuation rules, and ensuring there are no unnecessary repetitions),
        (iii) style (by ensuring the translations reflect the style of the source text and takes into account any cultural context),
        (iv) terminology (by ensuring terminology use is consistent and reflects the source text domain; and by only ensuring you use equivalent idioms {target_lang}).
        """),
        dedent(f"""
        Write a list of specific, helpful and constructive suggestions for improving the translation.
        Each suggestion should address one specific part of the translation.
        Output only the suggestions and nothing else.
        """)
    ],
    # output_model=Translation,
    show_tool_calls=False,
    read_tool_call_history=True,
    read_chat_history=True,
    debug_mode=False,
)

improvement_assistant = Assistant(
    name="Rewriter",
    role="Improve the translation based on the provided suggestions.",
    llm=OpenAIChat(model="gpt-4o", temperature=0, api_key=OPENAI_API_KEY),
    description=f"""
        You are an expert linguist, specializing in translation from {source_lang} to {target_lang}.
        You will be provided with a source text, translation and a list of suggestions.
    """,
    instructions=[
        f"""Your task is to carefully read, then edit, a translation from {source_lang} to {target_lang}, taking into account a list of expert suggestions and constructive criticisms.""",
        dedent(f"""
        Please take into account the expert suggestions when editing the translation. Edit the translation by ensuring:

        (i) accuracy (by correcting errors of addition, mistranslation, omission, or untranslated text),
        (ii) fluency (by applying {target_lang} grammar, spelling and punctuation rules and ensuring there are no unnecessary repetitions), \
        (iii) style (by ensuring the translations reflect the style of the source text)
        (iv) terminology (inappropriate for context, inconsistent use), or
        (v) other errors.
        """),
        "Output only the new translation and nothing else.",
    ],
    # output_model=Translation,
    show_tool_calls=False,
    read_tool_call_history=True,
    read_chat_history=True,
    debug_mode=False,
)

master_assistant = Assistant(
    name="Translation team",
    llm=OpenAIChat(model="gpt-4", temperature=0, api_key=OPENAI_API_KEY),
    team=[translation_assistant, suggestion_assistant, improvement_assistant],
    description=f"""You are an expert linguist, specializing in translation from {source_lang} to {target_lang}.""",
    instructions=[
        "Get an initial translation of the source text.",
        "Reflect on the initial translation and provide suggestions. Pass the suggestion to the next step.",
        "Write an improved translation based on the provided suggestions.",
    ],
    extra_instructions=[
        "Provide the source text, and latest translation in each step if available.",
    ],
    add_to_system_prompt="You will be given a list of assistants, each assistant should only be called once.",
    expected_output="Output only the new translation and nothing else.",
    read_tool_call_history=True,
    read_chat_history=True,
    debug_mode=True,
    tool_call_limit=40,
)

master_assistant.print_response(dedent(f"""
    Translate the following Hypebeast article from the {source_lang} to the {target_lang}.
    https://hypebeast.com/2024/5/mercer-labs-art-experiences-88rising-fam-1999-world
"""), markdown=True, stream=False)

[kayue]

The work around is to avoid using team. Just use multiple assistant with output model. Thanks.