fix(langchain): prefer structured tool inputs

langfuse/langchain/CallbackHandler.py

@@ -978,22 +978,25 @@ def on_tool_start(
             observation_type = self._get_observation_type_from_serialized(
                 serialized, "tool", **kwargs
             )
+            tool_input = kwargs.get("inputs")
+            if tool_input is None:
+                tool_input = input_str
 
             parent_observation = self._get_parent_observation(parent_run_id)
             if isinstance(parent_observation, Langfuse):
                 span = parent_observation.start_observation(
                     trace_context=self._trace_context,
                     name=self.get_langchain_run_name(serialized, **kwargs),
                     as_type=observation_type,
-                    input=input_str,
+                    input=tool_input,
                     metadata=meta,
                     level="DEBUG" if tags and LANGSMITH_TAG_HIDDEN in tags else None,
                 )
             else:
                 span = parent_observation.start_observation(
                     name=self.get_langchain_run_name(serialized, **kwargs),
                     as_type=observation_type,
-                    input=input_str,
+                    input=tool_input,
                     metadata=meta,
                     level="DEBUG" if tags and LANGSMITH_TAG_HIDDEN in tags else None,
                 )

tests/unit/test_langchain.py

@@ -791,6 +791,33 @@ class DummyControlFlowError(RuntimeError):
     assert not _has_run_state(handler, retriever_run_id)
 
 
+def test_tool_start_prefers_structured_inputs_when_available(
+    langfuse_memory_client, get_span, json_attr
+):
+    handler = CallbackHandler()
+    run_id = uuid4()
+    structured_inputs = {
+        "path": "/tmp/example.md",
+        "content": '# Title\n\nThis has "quotes" and JSON-like text: {"a": 1}',
+    }
+
+    handler.on_tool_start(
+        {"name": "write_document"},
+        str(structured_inputs),
+        run_id=run_id,
+        inputs=structured_inputs,
+    )
+    handler.on_tool_end("ok", run_id=run_id)
+
+    langfuse_memory_client.flush()
+    span = get_span("write_document")
+
+    assert (
+        json_attr(span, LangfuseOtelSpanAttributes.OBSERVATION_INPUT)
+        == structured_inputs
+    )
+
+
 def test_handled_tool_error_marks_observation_error(
     langfuse_memory_client, get_span, json_attr
 ):