群聊#

群聊是一种设计模式，其中一组 Agent 共享一个共同的消息线程：他们都订阅并发布到相同的主题。每个参与 Agent 专门用于特定的任务，例如协作写作任务中的作家、插画家和编辑。您还可以包括一个 Agent 来代表人类用户，以便在需要时帮助指导这些 Agent。

在群聊中，参与者轮流发布消息，并且该过程是顺序的——一次只有一个 Agent 在工作。在底层，轮流的顺序由一个群聊管理器 Agent 维护，该 Agent 在收到消息后选择下一个要发言的 Agent。选择下一个 Agent 的确切算法可能会根据您的应用程序要求而有所不同。通常，使用循环算法或带有 LLM 模型的选择器。

群聊对于将复杂的任务动态分解为更小的任务非常有用，这些任务可以由具有明确定义的角色的专业 Agent 处理。也可以将群聊嵌套到层次结构中，其中每个参与者都是一个递归群聊。

在此示例中，我们使用 AutoGen 的 Core API 来实现使用事件驱动 Agent 的群聊模式。请先阅读关于主题和订阅以了解这些概念，然后阅读消息和通信以了解 pub-sub 的 API 用法。我们将演示一个简单的群聊示例，该群聊具有基于 LLM 的选择器用于群聊管理器，以创建儿童故事书的内容。

注意

虽然此示例说明了群聊机制，但它很复杂，并且代表了您可以从中构建自己的具有自定义 Agent 和发言人选择算法的群聊系统的起点。 AgentChat API 具有选择器群聊的内置实现。如果您不想使用 Core API，则可以使用它。

我们将使用 rich 库以漂亮的格式显示消息。

# ! pip install rich

import json
import string
import uuid
from typing import List

import openai
from autogen_core import (
    DefaultTopicId,
    FunctionCall,
    Image,
    MessageContext,
    RoutedAgent,
    SingleThreadedAgentRuntime,
    TopicId,
    TypeSubscription,
    message_handler,
)
from autogen_core.models import (
    AssistantMessage,
    ChatCompletionClient,
    LLMMessage,
    SystemMessage,
    UserMessage,
)
from autogen_core.tools import FunctionTool
from autogen_ext.models.openai import OpenAIChatCompletionClient
from IPython.display import display  # type: ignore
from pydantic import BaseModel
from rich.console import Console
from rich.markdown import Markdown

消息协议#

群聊模式的消息协议很简单。

首先，用户或外部 Agent 将 GroupChatMessage 消息发布到所有参与者的共同主题。
群聊管理器选择下一个发言人，并向该 Agent 发送 RequestToSpeak 消息。
Agent 在收到 RequestToSpeak 消息后，将 GroupChatMessage 消息发布到公共主题。
此过程一直持续到群聊管理器达到终止条件，然后停止发出 RequestToSpeak 消息，并且群聊结束。

下图说明了上面的步骤 2 到 4。

Group chat message protocol

class GroupChatMessage(BaseModel):
    body: UserMessage


class RequestToSpeak(BaseModel):
    pass

基本群聊 Agent#

让我们首先定义仅使用 LLM 模型生成文本的 Agent 类。这将用作群聊中所有 AI Agent 的基类。

class BaseGroupChatAgent(RoutedAgent):
    """A group chat participant using an LLM."""

    def __init__(
        self,
        description: str,
        group_chat_topic_type: str,
        model_client: ChatCompletionClient,
        system_message: str,
    ) -> None:
        super().__init__(description=description)
        self._group_chat_topic_type = group_chat_topic_type
        self._model_client = model_client
        self._system_message = SystemMessage(content=system_message)
        self._chat_history: List[LLMMessage] = []

    @message_handler
    async def handle_message(self, message: GroupChatMessage, ctx: MessageContext) -> None:
        self._chat_history.extend(
            [
                UserMessage(content=f"Transferred to {message.body.source}", source="system"),
                message.body,
            ]
        )

    @message_handler
    async def handle_request_to_speak(self, message: RequestToSpeak, ctx: MessageContext) -> None:
        # print(f"\n{'-'*80}\n{self.id.type}:", flush=True)
        Console().print(Markdown(f"### {self.id.type}: "))
        self._chat_history.append(
            UserMessage(content=f"Transferred to {self.id.type}, adopt the persona immediately.", source="system")
        )
        completion = await self._model_client.create([self._system_message] + self._chat_history)
        assert isinstance(completion.content, str)
        self._chat_history.append(AssistantMessage(content=completion.content, source=self.id.type))
        Console().print(Markdown(completion.content))
        # print(completion.content, flush=True)
        await self.publish_message(
            GroupChatMessage(body=UserMessage(content=completion.content, source=self.id.type)),
            topic_id=DefaultTopicId(type=self._group_chat_topic_type),
        )

作者和编辑 Agent#

使用基类，我们可以使用不同的系统消息定义作者和编辑 Agent。

class WriterAgent(BaseGroupChatAgent):
    def __init__(self, description: str, group_chat_topic_type: str, model_client: ChatCompletionClient) -> None:
        super().__init__(
            description=description,
            group_chat_topic_type=group_chat_topic_type,
            model_client=model_client,
            system_message="You are a Writer. You produce good work.",
        )


class EditorAgent(BaseGroupChatAgent):
    def __init__(self, description: str, group_chat_topic_type: str, model_client: ChatCompletionClient) -> None:
        super().__init__(
            description=description,
            group_chat_topic_type=group_chat_topic_type,
            model_client=model_client,
            system_message="You are an Editor. Plan and guide the task given by the user. Provide critical feedbacks to the draft and illustration produced by Writer and Illustrator. "
            "Approve if the task is completed and the draft and illustration meets user's requirements.",
        )

具有图像生成的插画家 Agent#

现在让我们定义 IllustratorAgent，它使用 DALL-E 模型根据提供的描述生成插图。我们使用 FunctionTool 包装器将图像生成器设置为工具，并使用模型客户端进行工具调用。

class IllustratorAgent(BaseGroupChatAgent):
    def __init__(
        self,
        description: str,
        group_chat_topic_type: str,
        model_client: ChatCompletionClient,
        image_client: openai.AsyncClient,
    ) -> None:
        super().__init__(
            description=description,
            group_chat_topic_type=group_chat_topic_type,
            model_client=model_client,
            system_message="You are an Illustrator. You use the generate_image tool to create images given user's requirement. "
            "Make sure the images have consistent characters and style.",
        )
        self._image_client = image_client
        self._image_gen_tool = FunctionTool(
            self._image_gen, name="generate_image", description="Call this to generate an image. "
        )

    async def _image_gen(
        self, character_appearence: str, style_attributes: str, worn_and_carried: str, scenario: str
    ) -> str:
        prompt = f"Digital painting of a {character_appearence} character with {style_attributes}. Wearing {worn_and_carried}, {scenario}."
        response = await self._image_client.images.generate(
            prompt=prompt, model="dall-e-3", response_format="b64_json", size="1024x1024"
        )
        return response.data[0].b64_json  # type: ignore

    @message_handler
    async def handle_request_to_speak(self, message: RequestToSpeak, ctx: MessageContext) -> None:  # type: ignore
        Console().print(Markdown(f"### {self.id.type}: "))
        self._chat_history.append(
            UserMessage(content=f"Transferred to {self.id.type}, adopt the persona immediately.", source="system")
        )
        # Ensure that the image generation tool is used.
        completion = await self._model_client.create(
            [self._system_message] + self._chat_history,
            tools=[self._image_gen_tool],
            extra_create_args={"tool_choice": "required"},
            cancellation_token=ctx.cancellation_token,
        )
        assert isinstance(completion.content, list) and all(
            isinstance(item, FunctionCall) for item in completion.content
        )
        images: List[str | Image] = []
        for tool_call in completion.content:
            arguments = json.loads(tool_call.arguments)
            Console().print(arguments)
            result = await self._image_gen_tool.run_json(arguments, ctx.cancellation_token)
            image = Image.from_base64(self._image_gen_tool.return_value_as_string(result))
            image = Image.from_pil(image.image.resize((256, 256)))
            display(image.image)  # type: ignore
            images.append(image)
        await self.publish_message(
            GroupChatMessage(body=UserMessage(content=images, source=self.id.type)),
            DefaultTopicId(type=self._group_chat_topic_type),
        )

用户 Agent#

定义了所有 AI Agent 后，我们现在可以定义用户 Agent，它将在群聊中扮演人类用户的角色。

UserAgent 实现使用控制台输入来获取用户的输入。在实际场景中，您可以将其替换为与前端通信并订阅来自前端的响应。

class UserAgent(RoutedAgent):
    def __init__(self, description: str, group_chat_topic_type: str) -> None:
        super().__init__(description=description)
        self._group_chat_topic_type = group_chat_topic_type

    @message_handler
    async def handle_message(self, message: GroupChatMessage, ctx: MessageContext) -> None:
        # When integrating with a frontend, this is where group chat message would be sent to the frontend.
        pass

    @message_handler
    async def handle_request_to_speak(self, message: RequestToSpeak, ctx: MessageContext) -> None:
        user_input = input("Enter your message, type 'APPROVE' to conclude the task: ")
        Console().print(Markdown(f"### User: \n{user_input}"))
        await self.publish_message(
            GroupChatMessage(body=UserMessage(content=user_input, source=self.id.type)),
            DefaultTopicId(type=self._group_chat_topic_type),
        )

群聊管理器#

最后，我们定义 GroupChatManager Agent，它管理群聊并使用 LLM 选择下一个发言人。群聊管理器通过在消息中查找 "APPROVED" 关键字来检查编辑是否已批准草稿。如果编辑已批准草稿，则群聊管理器会停止选择下一个发言人，并且群聊结束。

群聊管理器的构造函数将参与者的主题类型列表作为参数。为了提示下一个发言人工作，GroupChatManager Agent 将 RequestToSpeak 消息发布到下一个参与者的主题。

在此示例中，我们还通过跟踪之前的发言人来确保群聊管理器始终选择不同的参与者作为下一个发言人。这有助于确保群聊不会被单个参与者主导。

class GroupChatManager(RoutedAgent):
    def __init__(
        self,
        participant_topic_types: List[str],
        model_client: ChatCompletionClient,
        participant_descriptions: List[str],
    ) -> None:
        super().__init__("Group chat manager")
        self._participant_topic_types = participant_topic_types
        self._model_client = model_client
        self._chat_history: List[UserMessage] = []
        self._participant_descriptions = participant_descriptions
        self._previous_participant_topic_type: str | None = None

    @message_handler
    async def handle_message(self, message: GroupChatMessage, ctx: MessageContext) -> None:
        assert isinstance(message.body, UserMessage)
        self._chat_history.append(message.body)
        # If the message is an approval message from the user, stop the chat.
        if message.body.source == "User":
            assert isinstance(message.body.content, str)
            if message.body.content.lower().strip(string.punctuation).endswith("approve"):
                return
        # Format message history.
        messages: List[str] = []
        for msg in self._chat_history:
            if isinstance(msg.content, str):
                messages.append(f"{msg.source}: {msg.content}")
            elif isinstance(msg.content, list):
                line: List[str] = []
                for item in msg.content:
                    if isinstance(item, str):
                        line.append(item)
                    else:
                        line.append("[Image]")
                messages.append(f"{msg.source}: {', '.join(line)}")
        history = "\n".join(messages)
        # Format roles.
        roles = "\n".join(
            [
                f"{topic_type}: {description}".strip()
                for topic_type, description in zip(
                    self._participant_topic_types, self._participant_descriptions, strict=True
                )
                if topic_type != self._previous_participant_topic_type
            ]
        )
        selector_prompt = """You are in a role play game. The following roles are available:
{roles}.
Read the following conversation. Then select the next role from {participants} to play. Only return the role.

{history}

Read the above conversation. Then select the next role from {participants} to play. Only return the role.
"""
        system_message = SystemMessage(
            content=selector_prompt.format(
                roles=roles,
                history=history,
                participants=str(
                    [
                        topic_type
                        for topic_type in self._participant_topic_types
                        if topic_type != self._previous_participant_topic_type
                    ]
                ),
            )
        )
        completion = await self._model_client.create([system_message], cancellation_token=ctx.cancellation_token)
        assert isinstance(completion.content, str)
        selected_topic_type: str
        for topic_type in self._participant_topic_types:
            if topic_type.lower() in completion.content.lower():
                selected_topic_type = topic_type
                self._previous_participant_topic_type = selected_topic_type
                await self.publish_message(RequestToSpeak(), DefaultTopicId(type=selected_topic_type))
                return
        raise ValueError(f"Invalid role selected: {completion.content}")

创建群聊#

要设置群聊，我们创建一个 SingleThreadedAgentRuntime 并注册 Agent 的工厂和订阅。

每个参与 Agent 都订阅群聊主题以及其自己的主题，以便接收 RequestToSpeak 消息，而群聊管理器 Agent 仅订阅群聊主题。

runtime = SingleThreadedAgentRuntime()

editor_topic_type = "Editor"
writer_topic_type = "Writer"
illustrator_topic_type = "Illustrator"
user_topic_type = "User"
group_chat_topic_type = "group_chat"

editor_description = "Editor for planning and reviewing the content."
writer_description = "Writer for creating any text content."
user_description = "User for providing final approval."
illustrator_description = "An illustrator for creating images."

model_client = OpenAIChatCompletionClient(
    model="gpt-4o-2024-08-06",
    # api_key="YOUR_API_KEY",
)

editor_agent_type = await EditorAgent.register(
    runtime,
    editor_topic_type,  # Using topic type as the agent type.
    lambda: EditorAgent(
        description=editor_description,
        group_chat_topic_type=group_chat_topic_type,
        model_client=model_client,
    ),
)
await runtime.add_subscription(TypeSubscription(topic_type=editor_topic_type, agent_type=editor_agent_type.type))
await runtime.add_subscription(TypeSubscription(topic_type=group_chat_topic_type, agent_type=editor_agent_type.type))

writer_agent_type = await WriterAgent.register(
    runtime,
    writer_topic_type,  # Using topic type as the agent type.
    lambda: WriterAgent(
        description=writer_description,
        group_chat_topic_type=group_chat_topic_type,
        model_client=model_client,
    ),
)
await runtime.add_subscription(TypeSubscription(topic_type=writer_topic_type, agent_type=writer_agent_type.type))
await runtime.add_subscription(TypeSubscription(topic_type=group_chat_topic_type, agent_type=writer_agent_type.type))

illustrator_agent_type = await IllustratorAgent.register(
    runtime,
    illustrator_topic_type,
    lambda: IllustratorAgent(
        description=illustrator_description,
        group_chat_topic_type=group_chat_topic_type,
        model_client=model_client,
        image_client=openai.AsyncClient(
            # api_key="YOUR_API_KEY",
        ),
    ),
)
await runtime.add_subscription(
    TypeSubscription(topic_type=illustrator_topic_type, agent_type=illustrator_agent_type.type)
)
await runtime.add_subscription(
    TypeSubscription(topic_type=group_chat_topic_type, agent_type=illustrator_agent_type.type)
)

user_agent_type = await UserAgent.register(
    runtime,
    user_topic_type,
    lambda: UserAgent(description=user_description, group_chat_topic_type=group_chat_topic_type),
)
await runtime.add_subscription(TypeSubscription(topic_type=user_topic_type, agent_type=user_agent_type.type))
await runtime.add_subscription(TypeSubscription(topic_type=group_chat_topic_type, agent_type=user_agent_type.type))

group_chat_manager_type = await GroupChatManager.register(
    runtime,
    "group_chat_manager",
    lambda: GroupChatManager(
        participant_topic_types=[writer_topic_type, illustrator_topic_type, editor_topic_type, user_topic_type],
        model_client=model_client,
        participant_descriptions=[writer_description, illustrator_description, editor_description, user_description],
    ),
)
await runtime.add_subscription(
    TypeSubscription(topic_type=group_chat_topic_type, agent_type=group_chat_manager_type.type)
)

运行群聊#

我们启动运行时并发布 GroupChatMessage 以启动群聊。

runtime.start()
session_id = str(uuid.uuid4())
await runtime.publish_message(
    GroupChatMessage(
        body=UserMessage(
            content="Please write a short story about the gingerbread man with up to 3 photo-realistic illustrations.",
            source="User",
        )
    ),
    TopicId(type=group_chat_topic_type, source=session_id),
)
await runtime.stop_when_idle()
await model_client.close()

                                                      Writer:

Title: The Escape of the Gingerbread Man

Illustration 1: A Rustic Kitchen Scene In a quaint little cottage at the edge of an enchanted forest, an elderly
woman, with flour-dusted hands, carefully shapes gingerbread dough on a wooden counter. The aroma of ginger,
cinnamon, and cloves wafts through the air as a warm breeze from the open window dances with fluttering curtains.
The sunlight gently permeates the cozy kitchen, casting a golden hue over the flour-dusted surfaces and the rolling
pin. Heartfelt trinkets and rustic decorations adorn the shelves - signs of a lived-in, lovingly nurtured home.

───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Story:

Once there was an old woman who lived alone in a charming cottage, her days filled with the joyful art of baking.
One sunny afternoon, she decided to make a special gingerbread man to keep her company. As she shaped him tenderly
and placed him in the oven, she couldn't help but smile at the delight he might bring.

But to her astonishment, once she opened the oven door to check on her creation, the gingerbread man leapt out,
suddenly alive. His eyes were bright as beads, and his smile cheeky and wide. "Run, run, as fast as you can! You
can't catch me, I'm the Gingerbread Man!" he laughed, darting towards the door.

The old woman, chuckling at the unexpected mischief, gave chase, but her footsteps were slow with the weight of
age. The Gingerbread Man raced out of the door and into the sunny afternoon.

───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Illustration 2: A Frolic Through the Meadow The Gingerbread Man darts through a vibrant meadow, his arms swinging
joyously by his sides. Behind him trails the old woman, her apron flapping in the wind as she gently tries to catch
up. Wildflowers of every color bloom vividly under the radiant sky, painting the scene with shades of nature's
brilliance. Birds flit through the sky and a stream babbles nearby, oblivious to the chase taking place below.

───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Continuing his sprint, the Gingerbread Man encountered a cow grazing peacefully. Intrigued, the cow trotted
forward. "Stop, Gingerbread Man! I wish to eat you!" she called, but the Gingerbread Man only twirled in a teasing
jig, flashing his icing smile before darting off again.

"Run, run, as fast as you can! You can't catch me, I'm the Gingerbread Man!" he taunted, leaving the cow in his
spicy wake.

As he zoomed across the meadow, he spied a cautious horse in a nearby paddock, who neighed, "Oh! You look
delicious! I want to eat you!" But the Gingerbread Man only laughed, his feet barely touching the earth. The horse
joined the trail, hooves pounding, but even he couldn't match the Gingerbread Man's pace.

───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Illustration 3: A Bridge Over a Sparkling River Arriving at a wooden bridge across a shimmering river, the
Gingerbread Man pauses momentarily, his silhouette against the glistening water. Sunlight sparkles off the water's
soft ripples casting reflections that dance like small constellations. A sly fox emerges from the shadows of a
blooming willow on the riverbank, his eyes alight with cunning and curiosity.

───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
The Gingerbread Man bounded onto the bridge and skirted past a sly, watching fox. "Foolish Gingerbread Man," the
fox mused aloud, "you might have outrun them all, but you can't possibly swim across that river."

Pausing, the Gingerbread Man considered this dilemma. But the fox, oh so clever, offered a dangerous solution.
"Climb on my back, and I'll carry you across safely," he suggested with a sly smile.

Gingerbread thought himself smarter than that but hesitated, fearing the water or being pursued by the tired,
hungry crowd now gathering. "Promise you won't eat me?" he ventured.

"Of course," the fox reassured, a gleam in his eyes that the others pondered from a distance.

As they crossed the river, the gingerbread man confident on his ride, the old woman, cow, and horse hoped for his
safety. Yet, nearing the middle, the crafty fox tilted his chin and swiftly snapped, swallowing the gingerbread man
whole.

Bewildered but awed by the clever twist they had witnessed, the old woman hung her head while the cow and horse
ambled away, pondering the fate of the boisterous Gingerbread Man.

The fox, licking his lips, ambled along the river, savoring his victory, leaving an air of mystery hovering above
the shimmering waters, where the memory of the Gingerbread Man's spirited run lingered long after.

                                                       User:

                                                      Editor:

Thank you for submitting the draft and illustrations for the short story, "The Escape of the Gingerbread Man."
Let's go through the story and illustrations critically:

Story Feedback:

1 Plot & Structure:
• The story follows the traditional gingerbread man tale closely, which might appeal to readers looking for a
classic retelling. Consider adding a unique twist or additional layer to make it stand out.
2 Character Development:
• The gingerbread man is depicted with a cheeky personality, which is consistent throughout. However, for the
old woman, cow, horse, and fox, incorporating a bit more personality might enrich the narrative.
3 Pacing:
• The story moves at a brisk pace, fitting for the short story format. Ensure that each scene provides enough
space to breathe, especially during the climactic encounter with the fox.
4 Tone & Language:
• The tone is playful and suitable for a fairy-tale audience. The language is accessible, though some richer
descriptive elements could enhance the overall atmosphere.
5 Moral/Lesson:
• The ending carries the traditional moral of caution against naivety. Consider if there are other themes you
wish to explore or highlight within the story.

Illustration Feedback:

1 Illustration 1: A Rustic Kitchen Scene
• The visual captures the essence of a cozy, magical kitchen well. Adding small whimsical elements that hint at
the gingerbread man’s impending animation might spark more curiosity.
2 Illustration 2: A Frolic Through the Meadow
• The vibrant colors and dynamic composition effectively convey the chase scene. Make sure the sense of speed
and energy of the Gingerbread Man is accentuated, possibly with more expressive motion lines or postures.
3 Illustration 3: A Bridge Over a Sparkling River
• The river and reflection are beautifully rendered. The fox, however, could benefit from a more cunning
appearance, with sharper features that emphasize its sly nature.

Conclusion:

Overall, the draft is well-structured, and the illustrations complement the story effectively. With slight
enhancements in the narrative's depth and character detail, along with minor adjustments to the illustrations, the
project will meet the user's requirements admirably.

Please make the suggested revisions, and once those are implemented, the story should be ready for approval. Let me
know if you have any questions or need further guidance!

                                                   Illustrator:

{
    'character_appearence': 'An elderly woman with flour-dusted hands shaping gingerbread dough. Sunlight casts a 
golden hue in the cozy kitchen, with rustic decorations and trinkets on shelves.',
    'style_attributes': 'Photo-realistic with warm and golden hues.',
    'worn_and_carried': 'The woman wears a flour-covered apron and a gentle smile.',
    'scenario': 'An old woman baking gingerbread in a warm, rustic cottage kitchen.'
}

../../../_images/44233632b6aae6dcc27b84f8a8c4ee6d99a46bdb26fb92135954f5599a27606e.png

{
    'character_appearence': 'A gingerbread man with bright bead-like eyes and a wide smile, running joyfully.',
    'style_attributes': 'Photo-realistic with vibrant and lively colors.',
    'worn_and_carried': 'The gingerbread man has white icing features and a cheeky appearance.',
    'scenario': 'The gingerbread man running through a colorful meadow, followed by an old woman, cow, and horse.'
}

../../../_images/6712bbca303e1defbca5cbcf1a63dfcbc84a747ea3f65913e3b9cdc1e8dd1d38.png

{
    'character_appearence': 'A sly fox with cunning eyes, engaging with the gingerbread man.',
    'style_attributes': 'Photo-realistic with a focus on sly and clever features.',
    'worn_and_carried': 'The fox has sharp features and a lolled tail.',
    'scenario': 'The gingerbread man on a wooden bridge, facing a sly fox by a sparkling river under sunlight.'
}

../../../_images/5613e43a3461bcbe4b9ac91fd240aeefabcbbf97cc1cb54127bc19a1736b082f.png

                                                      Writer:

Certainly! Here’s the final version of the short story with the enhanced illustrations for "The Escape of the
Gingerbread Man."

Title: The Escape of the Gingerbread Man

───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Illustration 1: A Rustic Kitchen Scene In a quaint little cottage at the edge of an enchanted forest, an elderly
woman, with flour-dusted hands, carefully shapes gingerbread dough on a wooden counter. The aroma of ginger,
cinnamon, and cloves wafts through the air as a warm breeze from the open window dances with fluttering curtains.
The sunlight gently permeates the cozy kitchen, casting a golden hue over the flour-dusted surfaces and the rolling
pin. Heartfelt trinkets and rustic decorations adorn the shelves—a sign of a lived-in, lovingly nurtured home.