Declarative BaseChat Agents (#5055)

* v1, make assistant agent declarative

* make head tail context declarative

* update and formatting

* update assistant, format updates

* make websurfer declarative

* update formatting

* move declarative docs to advanced section

* remove tools until implemented

* minor updates to termination conditions

* update docs

python/packages/autogen-agentchat/src/autogen_agentchat/agents/_assistant_agent.py

@@ -13,7 +13,7 @@
     Sequence,
 )
 
-from autogen_core import CancellationToken, FunctionCall
+from autogen_core import CancellationToken, Component, ComponentModel, FunctionCall
 from autogen_core.memory import Memory
 from autogen_core.model_context import (
     ChatCompletionContext,
@@ -28,6 +28,8 @@
     UserMessage,
 )
 from autogen_core.tools import FunctionTool, Tool
+from pydantic import BaseModel
+from typing_extensions import Self
 
 from .. import EVENT_LOGGER_NAME
 from ..base import Handoff as HandoffBase
@@ -49,7 +51,21 @@
 event_logger = logging.getLogger(EVENT_LOGGER_NAME)
 
 
-class AssistantAgent(BaseChatAgent):
+class AssistantAgentConfig(BaseModel):
+    """The declarative configuration for the assistant agent."""
+
+    name: str
+    model_client: ComponentModel
+    # tools: List[Any] | None = None # TBD
+    handoffs: List[HandoffBase | str] | None = None
+    model_context: ComponentModel | None = None
+    description: str
+    system_message: str | None = None
+    reflect_on_tool_use: bool
+    tool_call_summary_format: str
+
+
+class AssistantAgent(BaseChatAgent, Component[AssistantAgentConfig]):
     """An agent that provides assistance with tool use.
 
     The :meth:`on_messages` returns a :class:`~autogen_agentchat.base.Response`
@@ -229,6 +245,9 @@ async def main() -> None:
             See `o1 beta limitations <https://platform.openai.com/docs/guides/reasoning#beta-limitations>`_ for more details.
     """
 
+    component_config_schema = AssistantAgentConfig
+    component_provider_override = "autogen_agentchat.agents.AssistantAgent"
+
     def __init__(
         self,
         name: str,
@@ -462,3 +481,40 @@ async def load_state(self, state: Mapping[str, Any]) -> None:
         assistant_agent_state = AssistantAgentState.model_validate(state)
         # Load the model context state.
         await self._model_context.load_state(assistant_agent_state.llm_context)
+
+    def _to_config(self) -> AssistantAgentConfig:
+        """Convert the assistant agent to a declarative config."""
+
+        # raise an error if tools is not empty until it is implemented
+        # TBD : Implement serializing tools and remove this check.
+        if self._tools and len(self._tools) > 0:
+            raise NotImplementedError("Serializing tools is not implemented yet.")
+
+        return AssistantAgentConfig(
+            name=self.name,
+            model_client=self._model_client.dump_component(),
+            # tools=[], # TBD
+            handoffs=list(self._handoffs.values()),
+            model_context=self._model_context.dump_component(),
+            description=self.description,
+            system_message=self._system_messages[0].content
+            if self._system_messages and isinstance(self._system_messages[0].content, str)
+            else None,
+            reflect_on_tool_use=self._reflect_on_tool_use,
+            tool_call_summary_format=self._tool_call_summary_format,
+        )
+
+    @classmethod
+    def _from_config(cls, config: AssistantAgentConfig) -> Self:
+        """Create an assistant agent from a declarative config."""
+        return cls(
+            name=config.name,
+            model_client=ChatCompletionClient.load_component(config.model_client),
+            # tools=[], # TBD
+            handoffs=config.handoffs,
+            model_context=None,
+            description=config.description,
+            system_message=config.system_message,
+            reflect_on_tool_use=config.reflect_on_tool_use,
+            tool_call_summary_format=config.tool_call_summary_format,
+        )

python/packages/autogen-agentchat/src/autogen_agentchat/agents/_base_chat_agent.py

@@ -1,7 +1,8 @@
 from abc import ABC, abstractmethod
 from typing import Any, AsyncGenerator, List, Mapping, Sequence
 
-from autogen_core import CancellationToken
+from autogen_core import CancellationToken, ComponentBase
+from pydantic import BaseModel
 
 from ..base import ChatAgent, Response, TaskResult
 from ..messages import (
@@ -13,7 +14,7 @@
 from ..state import BaseState
 
 
-class BaseChatAgent(ChatAgent, ABC):
+class BaseChatAgent(ChatAgent, ABC, ComponentBase[BaseModel]):
     """Base class for a chat agent.
 
     This abstract class provides a base implementation for a :class:`ChatAgent`.
@@ -35,6 +36,8 @@ class BaseChatAgent(ChatAgent, ABC):
         This design principle must be followed when creating a new agent.
     """
 
+    component_type = "agent"
+
     def __init__(self, name: str, description: str) -> None:
         self._name = name
         if self._name.isidentifier() is False:

python/packages/autogen-agentchat/src/autogen_agentchat/agents/_user_proxy_agent.py

@@ -5,7 +5,9 @@
 from inspect import iscoroutinefunction
 from typing import Any, AsyncGenerator, Awaitable, Callable, ClassVar, Generator, Optional, Sequence, Union, cast
 
-from autogen_core import CancellationToken
+from autogen_core import CancellationToken, Component
+from pydantic import BaseModel
+from typing_extensions import Self
 
 from ..base import Response
 from ..messages import AgentEvent, ChatMessage, HandoffMessage, TextMessage, UserInputRequestedEvent
@@ -24,7 +26,15 @@ async def cancellable_input(prompt: str, cancellation_token: Optional[Cancellati
     return await task
 
 
-class UserProxyAgent(BaseChatAgent):
+class UserProxyAgentConfig(BaseModel):
+    """Declarative configuration for the UserProxyAgent."""
+
+    name: str
+    description: str = "A human user"
+    input_func: str | None = None
+
+
+class UserProxyAgent(BaseChatAgent, Component[UserProxyAgentConfig]):
     """An agent that can represent a human user through an input function.
 
     This agent can be used to represent a human user in a chat system by providing a custom input function.
@@ -109,6 +119,10 @@ async def cancellable_user_agent():
                     print(f"BaseException: {e}")
     """
 
+    component_type = "agent"
+    component_provider_override = "autogen_agentchat.agents.UserProxyAgent"
+    component_config_schema = UserProxyAgentConfig
+
     class InputRequestContext:
         def __init__(self) -> None:
             raise RuntimeError(
@@ -218,3 +232,11 @@ async def on_messages_stream(
     async def on_reset(self, cancellation_token: Optional[CancellationToken] = None) -> None:
         """Reset agent state."""
         pass
+
+    def _to_config(self) -> UserProxyAgentConfig:
+        # TODO: Add ability to serialie input_func
+        return UserProxyAgentConfig(name=self.name, description=self.description, input_func=None)
+
+    @classmethod
+    def _from_config(cls, config: UserProxyAgentConfig) -> Self:
+        return cls(name=config.name, description=config.description, input_func=None)

python/packages/autogen-agentchat/src/autogen_agentchat/base/_termination.py

@@ -48,7 +48,6 @@ async def main() -> None:
     """
 
     component_type = "termination"
-    # component_config_schema = BaseModel  # type: ignore
 
     @property
     @abstractmethod

python/packages/autogen-agentchat/src/autogen_agentchat/conditions/_terminations.py

@@ -16,7 +16,6 @@ class StopMessageTerminationConfig(BaseModel):
 class StopMessageTermination(TerminationCondition, Component[StopMessageTerminationConfig]):
     """Terminate the conversation if a StopMessage is received."""
 
-    component_type = "termination"
     component_config_schema = StopMessageTerminationConfig
     component_provider_override = "autogen_agentchat.conditions.StopMessageTermination"
 
@@ -58,7 +57,6 @@ class MaxMessageTermination(TerminationCondition, Component[MaxMessageTerminatio
         max_messages: The maximum number of messages allowed in the conversation.
     """
 
-    component_type = "termination"
     component_config_schema = MaxMessageTerminationConfig
     component_provider_override = "autogen_agentchat.conditions.MaxMessageTermination"
 
@@ -104,7 +102,6 @@ class TextMentionTermination(TerminationCondition, Component[TextMentionTerminat
         text: The text to look for in the messages.
     """
 
-    component_type = "termination"
     component_config_schema = TextMentionTerminationConfig
     component_provider_override = "autogen_agentchat.conditions.TextMentionTermination"
 
@@ -159,7 +156,6 @@ class TokenUsageTermination(TerminationCondition, Component[TokenUsageTerminatio
         ValueError: If none of max_total_token, max_prompt_token, or max_completion_token is provided.
     """
 
-    component_type = "termination"
     component_config_schema = TokenUsageTerminationConfig
     component_provider_override = "autogen_agentchat.conditions.TokenUsageTermination"
 
@@ -234,7 +230,6 @@ class HandoffTermination(TerminationCondition, Component[HandoffTerminationConfi
         target (str): The target of the handoff message.
     """
 
-    component_type = "termination"
     component_config_schema = HandoffTerminationConfig
     component_provider_override = "autogen_agentchat.conditions.HandoffTermination"
 
@@ -279,7 +274,6 @@ class TimeoutTermination(TerminationCondition, Component[TimeoutTerminationConfi
         timeout_seconds: The maximum duration in seconds before terminating the conversation.
     """
 
-    component_type = "termination"
     component_config_schema = TimeoutTerminationConfig
     component_provider_override = "autogen_agentchat.conditions.TimeoutTermination"
 
@@ -339,7 +333,6 @@ class ExternalTermination(TerminationCondition, Component[ExternalTerminationCon
 
     """
 
-    component_type = "termination"
     component_config_schema = ExternalTerminationConfig
     component_provider_override = "autogen_agentchat.conditions.ExternalTermination"
 
@@ -389,7 +382,6 @@ class SourceMatchTermination(TerminationCondition, Component[SourceMatchTerminat
         TerminatedException: If the termination condition has already been reached.
     """
 
-    component_type = "termination"
     component_config_schema = SourceMatchTerminationConfig
     component_provider_override = "autogen_agentchat.conditions.SourceMatchTermination"
 

python/packages/autogen-agentchat/tests/test_assistant_agent.py

@@ -592,3 +592,51 @@ class BadMemory:
 
     assert not isinstance(BadMemory(), Memory)
     assert isinstance(ListMemory(), Memory)
+
+
+@pytest.mark.asyncio
+async def test_assistant_agent_declarative(monkeypatch: pytest.MonkeyPatch) -> None:
+    model = "gpt-4o-2024-05-13"
+    chat_completions = [
+        ChatCompletion(
+            id="id1",
+            choices=[
+                Choice(
+                    finish_reason="stop",
+                    index=0,
+                    message=ChatCompletionMessage(content="Response to message 3", role="assistant"),
+                )
+            ],
+            created=0,
+            model=model,
+            object="chat.completion",
+            usage=CompletionUsage(prompt_tokens=10, completion_tokens=5, total_tokens=15),
+        ),
+    ]
+    mock = _MockChatCompletion(chat_completions)
+    monkeypatch.setattr(AsyncCompletions, "create", mock.mock_create)
+    model_context = BufferedChatCompletionContext(buffer_size=2)
+    agent = AssistantAgent(
+        "test_agent",
+        model_client=OpenAIChatCompletionClient(model=model, api_key=""),
+        model_context=model_context,
+    )
+
+    agent_config = agent.dump_component()
+    assert agent_config.provider == "autogen_agentchat.agents.AssistantAgent"
+
+    agent2 = AssistantAgent.load_component(agent_config)
+    assert agent2.name == agent.name
+
+    agent3 = AssistantAgent(
+        "test_agent",
+        model_client=OpenAIChatCompletionClient(model=model, api_key=""),
+        model_context=model_context,
+        tools=[
+            _pass_function,
+            _fail_function,
+            FunctionTool(_echo_function, description="Echo"),
+        ],
+    )
+    with pytest.raises(NotImplementedError):
+        agent3.dump_component()

python/packages/autogen-agentchat/tests/test_declarative_components.py

@@ -11,6 +11,11 @@
     TokenUsageTermination,
 )
 from autogen_core import ComponentLoader, ComponentModel
+from autogen_core.model_context import (
+    BufferedChatCompletionContext,
+    HeadAndTailChatCompletionContext,
+    UnboundedChatCompletionContext,
+)
 
 
 @pytest.mark.asyncio
@@ -92,3 +97,35 @@ async def test_termination_declarative() -> None:
     # Test loading complex composition
     loaded_composite = ComponentLoader.load_component(composite_config)
     assert isinstance(loaded_composite, AndTerminationCondition)
+
+
+@pytest.mark.asyncio
+async def test_chat_completion_context_declarative() -> None:
+    unbounded_context = UnboundedChatCompletionContext()
+    buffered_context = BufferedChatCompletionContext(buffer_size=5)
+    head_tail_context = HeadAndTailChatCompletionContext(head_size=3, tail_size=2)
+
+    # Test serialization
+    unbounded_config = unbounded_context.dump_component()
+    assert unbounded_config.provider == "autogen_core.model_context.UnboundedChatCompletionContext"
+
+    buffered_config = buffered_context.dump_component()
+    assert buffered_config.provider == "autogen_core.model_context.BufferedChatCompletionContext"
+    assert buffered_config.config["buffer_size"] == 5
+
+    head_tail_config = head_tail_context.dump_component()
+    assert head_tail_config.provider == "autogen_core.model_context.HeadAndTailChatCompletionContext"
+    assert head_tail_config.config["head_size"] == 3
+    assert head_tail_config.config["tail_size"] == 2
+
+    # Test deserialization
+    loaded_unbounded = ComponentLoader.load_component(unbounded_config, UnboundedChatCompletionContext)
+    assert isinstance(loaded_unbounded, UnboundedChatCompletionContext)
+
+    loaded_buffered = ComponentLoader.load_component(buffered_config, BufferedChatCompletionContext)
+
+    assert isinstance(loaded_buffered, BufferedChatCompletionContext)
+
+    loaded_head_tail = ComponentLoader.load_component(head_tail_config, HeadAndTailChatCompletionContext)
+
+    assert isinstance(loaded_head_tail, HeadAndTailChatCompletionContext)

python/packages/autogen-core/docs/src/user-guide/agentchat-user-guide/index.md

@@ -66,6 +66,18 @@ Sample code and use cases
 
 How to migrate from AutoGen 0.2.x to 0.4.x.
 :::
+
+:::{grid-item-card} {fas}`save;pst-color-primary` Serialize Components
+:link: ./serialize-components.html
+
+Serialize and deserialize components
+:::
+
+:::{grid-item-card} {fas}`brain;pst-color-primary` Memory
+:link: ./memory.html
+
+Add memory capabilities to your agents
+:::
 ::::
 
 ```{toctree}
@@ -91,8 +103,7 @@ tutorial/human-in-the-loop
 tutorial/termination
 tutorial/custom-agents
 tutorial/state
-tutorial/declarative
-tutorial/memory
+
 ```
 
 ```{toctree}
@@ -103,6 +114,8 @@ tutorial/memory
 selector-group-chat
 swarm
 magentic-one
+memory
+serialize-components
 ```
 
 ```{toctree}