rfc: trace standard tool schema directly

libs/core/langchain_core/callbacks/base.py

@@ -257,6 +257,7 @@ class CallbackManagerMixin:
         parent_run_id: Optional[UUID] = None,
         tags: Optional[list[str]] = None,
         metadata: Optional[dict[str, Any]] = None,
+        tools: Optional[Sequence[_ToolSchema]] = None,
         **kwargs: Any,
     ) -> Any:
         """Run when a chat model starts running.
@@ -1070,3 +1071,11 @@ class BaseCallbackManager(CallbackManagerMixin):
 
 
 Callbacks = Optional[Union[list[BaseCallbackHandler], BaseCallbackManager]]
+
+
+from typing_extensions import TypedDict
+
+class _ToolSchema(TypedDict):
+    name: str
+    description: str
+    parameters: dict

libs/core/langchain_core/language_models/chat_models.py

@@ -272,6 +272,16 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
             )
             raise ValueError(msg)
 
+    def _standardize_tools(self, tools: Optional[Sequence]) -> Optional[List[_ToolSchema]]:
+        """Convert tools to standard format for tracing."""
+        if not tools:
+            return tools
+
+        try:
+            return [convert_to_openai_tool(tool) for tool in tools]
+        except Exception:
+            return None
+
     def invoke(
         self,
         input: LanguageModelInput,
@@ -357,6 +367,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
         config: Optional[RunnableConfig] = None,
         *,
         stop: Optional[list[str]] = None,
+        tools: Optional[Sequence] = None,
         **kwargs: Any,
     ) -> Iterator[BaseMessageChunk]:
         if not self._should_stream(async_api=False, **{**kwargs, "stream": True}):
@@ -367,6 +378,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
         else:
             config = ensure_config(config)
             messages = self._convert_input(input).to_messages()
+            tools_to_trace = self._standardize_tools(tools)
             params = self._get_invocation_params(stop=stop, **kwargs)
             options = {"stop": stop, **kwargs}
             inheritable_metadata = {
@@ -390,6 +402,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                 name=config.get("run_name"),
                 run_id=config.pop("run_id", None),
                 batch_size=1,
+                tools=tools_to_trace,
             )
             generation: Optional[ChatGenerationChunk] = None
 

libs/core/langchain_core/tracers/langchain.py

@@ -7,7 +7,7 @@ import logging
 import warnings
 from concurrent.futures import ThreadPoolExecutor
 from datetime import datetime, timezone
-from typing import TYPE_CHECKING, Any, Optional, Union
+from typing import TYPE_CHECKING, Any, Optional, Union, Sequence
 from uuid import UUID
 
 from langsmith import Client
@@ -21,6 +21,7 @@ from tenacity import (
     wait_exponential_jitter,
 )
 
+from langchain_core.callbacks.base import _ToolSchema
 from langchain_core.env import get_runtime_environment
 from langchain_core.load import dumpd
 from langchain_core.outputs import ChatGenerationChunk, GenerationChunk
@@ -135,6 +136,7 @@ class LangChainTracer(BaseTracer):
         parent_run_id: Optional[UUID] = None,
         metadata: Optional[dict[str, Any]] = None,
         name: Optional[str] = None,
+        tools: Optional[Sequence[_ToolSchema]] = None,
         **kwargs: Any,
     ) -> Run:
         """Start a trace for an LLM run.
@@ -147,6 +149,7 @@ class LangChainTracer(BaseTracer):
             parent_run_id: The parent run ID. Defaults to None.
             metadata: The metadata. Defaults to None.
             name: The name. Defaults to None.
+            tools: The tools.
             kwargs: Additional keyword arguments.
 
         Returns:
@@ -155,11 +158,14 @@ class LangChainTracer(BaseTracer):
         start_time = datetime.now(timezone.utc)
         if metadata:
             kwargs.update({"metadata": metadata})
+        inputs = {"messages": [[dumpd(msg) for msg in batch] for batch in messages]}
+        if tools:
+            inputs["tools"] = tools
         chat_model_run = Run(
             id=run_id,
             parent_run_id=parent_run_id,
             serialized=serialized,
-            inputs={"messages": [[dumpd(msg) for msg in batch] for batch in messages]},
+            inputs=inputs,
             extra=kwargs,
             events=[{"name": "start", "time": start_time}],
             start_time=start_time,