diff --git a/src/google/adk/models/anthropic_llm.py b/src/google/adk/models/anthropic_llm.py
index a14c767f23..db8a85099c 100644
--- a/src/google/adk/models/anthropic_llm.py
+++ b/src/google/adk/models/anthropic_llm.py
@@ -124,20 +124,26 @@ def part_to_message_block(
     content = ""
     response_data = part.function_response.response
 
-    # Handle response with content array
-    if "content" in response_data and response_data["content"]:
-      content_items = []
-      for item in response_data["content"]:
-        if isinstance(item, dict):
-          # Handle text content blocks
-          if item.get("type") == "text" and "text" in item:
-            content_items.append(item["text"])
+    # Handle response with content payloads.
+    if "content" in response_data:
+      content_value = response_data["content"]
+      if isinstance(content_value, list):
+        content_items = []
+        for item in content_value:
+          if isinstance(item, dict):
+            # Handle text content blocks
+            if item.get("type") == "text" and "text" in item:
+              content_items.append(item["text"])
+            else:
+              # Handle other structured content
+              content_items.append(str(item))
           else:
-            # Handle other structured content
             content_items.append(str(item))
-        else:
-          content_items.append(str(item))
-      content = "\n".join(content_items) if content_items else ""
+        content = "\n".join(content_items) if content_items else ""
+      elif isinstance(content_value, dict):
+        content = json.dumps(content_value)
+      elif content_value is not None:
+        content = str(content_value)
     # We serialize to str here
     # SDK ref: anthropic.types.tool_result_block_param
     # https://github.com/anthropics/anthropic-sdk-python/blob/main/src/anthropic/types/tool_result_block_param.py
diff --git a/tests/unittests/models/test_anthropic_llm.py b/tests/unittests/models/test_anthropic_llm.py
index fb44d5c8e7..520d1eb216 100644
--- a/tests/unittests/models/test_anthropic_llm.py
+++ b/tests/unittests/models/test_anthropic_llm.py
@@ -776,6 +776,52 @@ def test_part_to_message_block_with_multiple_content_items():
   assert result["content"] == "First part\nSecond part"
 
 
+def test_part_to_message_block_with_string_content():
+  """String content should stay intact instead of being split per character."""
+  from google.adk.models.anthropic_llm import part_to_message_block
+
+  response_part = types.Part.from_function_response(
+      name="load_skill_resource",
+      response={
+          "skill_name": "my_skill",
+          "file_path": "references/doc.md",
+          "content": "Hello",
+      },
+  )
+  response_part.function_response.id = "test_id_790"
+
+  result = part_to_message_block(response_part)
+
+  assert isinstance(result, dict)
+  assert result["tool_use_id"] == "test_id_790"
+  assert result["type"] == "tool_result"
+  assert not result["is_error"]
+  assert result["content"] == "Hello"
+
+
+def test_part_to_message_block_with_empty_string_content():
+  """An explicit empty content payload should stay empty."""
+  from google.adk.models.anthropic_llm import part_to_message_block
+
+  response_part = types.Part.from_function_response(
+      name="load_skill_resource",
+      response={
+          "skill_name": "my_skill",
+          "file_path": "references/doc.md",
+          "content": "",
+      },
+  )
+  response_part.function_response.id = "test_id_791"
+
+  result = part_to_message_block(response_part)
+
+  assert isinstance(result, dict)
+  assert result["tool_use_id"] == "test_id_791"
+  assert result["type"] == "tool_result"
+  assert not result["is_error"]
+  assert result["content"] == ""
+
+
 def test_part_to_message_block_with_pdf_document():
   """Test that part_to_message_block handles PDF document parts."""
   pdf_data = b"%PDF-1.4 fake pdf content"