core[patch]: support final AIMessage responses in `tool_example_to_me…

…ssages` (#28267)

We have a test
[test_structured_few_shot_examples](https://github.com/langchain-ai/langchain/blob/ad4333ca032033097c663dfe818c5c892c368bd6/libs/standard-tests/langchain_tests/integration_tests/chat_models.py#L546)
in standard integration tests that implements a version of tool-calling
few shot examples that works with ~all tested providers. The formulation
supported by ~all providers is: `human message, tool call, tool message,
AI reponse`.

Here we update
`langchain_core.utils.function_calling.tool_example_to_messages` to
support this formulation.

The `tool_example_to_messages` util is undocumented outside of our API
reference. IMO, if we are testing that this function works across all
providers, it can be helpful to feature it in our guides. The structured
few-shot examples we document at the moment require users to implement
this function and can be simplified.

libs/core/langchain_core/utils/function_calling.py

@@ -22,7 +22,7 @@
 from pydantic import BaseModel
 from typing_extensions import TypedDict, get_args, get_origin, is_typeddict
 
-from langchain_core._api import deprecated
+from langchain_core._api import beta, deprecated
 from langchain_core.messages import AIMessage, BaseMessage, HumanMessage, ToolMessage
 from langchain_core.utils.json_schema import dereference_refs
 from langchain_core.utils.pydantic import is_basemodel_subclass
@@ -494,21 +494,28 @@ def convert_to_openai_tool(
     return {"type": "function", "function": oai_function}
 
 
+@beta()
 def tool_example_to_messages(
-    input: str, tool_calls: list[BaseModel], tool_outputs: Optional[list[str]] = None
+    input: str,
+    tool_calls: list[BaseModel],
+    tool_outputs: Optional[list[str]] = None,
+    *,
+    ai_response: Optional[str] = None,
 ) -> list[BaseMessage]:
     """Convert an example into a list of messages that can be fed into an LLM.
 
     This code is an adapter that converts a single example to a list of messages
     that can be fed into a chat model.
 
-    The list of messages per example corresponds to:
+    The list of messages per example by default corresponds to:
 
     1) HumanMessage: contains the content from which content should be extracted.
     2) AIMessage: contains the extracted information from the model
     3) ToolMessage: contains confirmation to the model that the model requested a tool
         correctly.
 
+    If `ai_response` is specified, there will be a final AIMessage with that response.
+
     The ToolMessage is required because some chat models are hyper-optimized for agents
     rather than for an extraction use case.
 
@@ -519,6 +526,7 @@ def tool_example_to_messages(
         tool_outputs: Optional[List[str]], a list of tool call outputs.
             Does not need to be provided. If not provided, a placeholder value
             will be inserted. Defaults to None.
+        ai_response: Optional[str], if provided, content for a final AIMessage.
 
     Returns:
         A list of messages
@@ -584,6 +592,9 @@ class Person(BaseModel):
     )
     for output, tool_call_dict in zip(tool_outputs, openai_tool_calls):
         messages.append(ToolMessage(content=output, tool_call_id=tool_call_dict["id"]))  # type: ignore
+
+    if ai_response:
+        messages.append(AIMessage(content=ai_response))
     return messages
 
 

libs/core/tests/unit_tests/utils/test_function_calling.py

@@ -679,6 +679,24 @@ def test_tool_outputs() -> None:
     ]
     assert messages[2].content == "Output1"
 
+    # Test final AI response
+    messages = tool_example_to_messages(
+        input="This is an example",
+        tool_calls=[
+            FakeCall(data="ToolCall1"),
+        ],
+        tool_outputs=["Output1"],
+        ai_response="The output is Output1",
+    )
+    assert len(messages) == 4
+    assert isinstance(messages[0], HumanMessage)
+    assert isinstance(messages[1], AIMessage)
+    assert isinstance(messages[2], ToolMessage)
+    assert isinstance(messages[3], AIMessage)
+    response = messages[3]
+    assert response.content == "The output is Output1"
+    assert not response.tool_calls
+
 
 @pytest.mark.parametrize("use_extension_typed_dict", [True, False])
 @pytest.mark.parametrize("use_extension_annotated", [True, False])

libs/standard-tests/langchain_tests/integration_tests/chat_models.py

@@ -17,6 +17,7 @@
 from langchain_core.output_parsers import StrOutputParser
 from langchain_core.prompts import ChatPromptTemplate
 from langchain_core.tools import tool
+from langchain_core.utils.function_calling import tool_example_to_messages
 from pydantic import BaseModel, Field
 from pydantic.v1 import BaseModel as BaseModelV1
 from pydantic.v1 import Field as FieldV1
@@ -857,33 +858,20 @@ def test_structured_few_shot_examples(self, model: BaseChatModel) -> None:
         if not self.has_tool_calling:
             pytest.skip("Test requires tool calling.")
         model_with_tools = model.bind_tools([my_adder_tool], tool_choice="any")
-        function_name = "my_adder_tool"
-        function_args = {"a": 1, "b": 2}
         function_result = json.dumps({"result": 3})
 
-        messages_string_content = [
-            HumanMessage("What is 1 + 2"),
-            AIMessage(
-                "",
-                tool_calls=[
-                    {
-                        "name": function_name,
-                        "args": function_args,
-                        "id": "abc123",
-                        "type": "tool_call",
-                    },
-                ],
-            ),
-            ToolMessage(
-                function_result,
-                name=function_name,
-                tool_call_id="abc123",
-            ),
-            AIMessage(function_result),
-            HumanMessage("What is 3 + 4"),
-        ]
-        result_string_content = model_with_tools.invoke(messages_string_content)
-        assert isinstance(result_string_content, AIMessage)
+        tool_schema = my_adder_tool.args_schema
+        assert tool_schema is not None
+        few_shot_messages = tool_example_to_messages(
+            "What is 1 + 2",
+            [tool_schema(a=1, b=2)],
+            tool_outputs=[function_result],
+            ai_response=function_result,
+        )
+
+        messages = few_shot_messages + [HumanMessage("What is 3 + 4")]
+        result = model_with_tools.invoke(messages)
+        assert isinstance(result, AIMessage)
 
     def test_image_inputs(self, model: BaseChatModel) -> None:
         if not self.supports_image_inputs: