fix: anthropic - connect and enrich finish reason support

Author: sebastienc

src/google/adk/models/anthropic_llm.py

@@ -133,10 +133,12 @@ def to_claude_role(role: Optional[str]) -> Literal["user", "assistant"]:
 def to_google_genai_finish_reason(
     anthropic_stop_reason: Optional[str],
 ) -> types.FinishReason:
-  if anthropic_stop_reason in ["end_turn", "stop_sequence", "tool_use"]:
+  if anthropic_stop_reason in ["end_turn", "stop_sequence", "tool_use", "pause_turn"]:
     return "STOP"
   if anthropic_stop_reason == "max_tokens":
     return "MAX_TOKENS"
+  if anthropic_stop_reason == "refusal":
+    return "SAFETY"
   return "FINISH_REASON_UNSPECIFIED"
 
 
@@ -343,8 +345,7 @@ def message_to_generate_content_response(
               message.usage.input_tokens + message.usage.output_tokens
           ),
       ),
-      # TODO: Deal with these later.
-      # finish_reason=to_google_genai_finish_reason(message.stop_reason),
+      finish_reason=to_google_genai_finish_reason(message.stop_reason),
   )
 
 
@@ -547,6 +548,7 @@ async def _generate_content_streaming(
     redacted_thinking_blocks: dict[int, str] = {}
     input_tokens = 0
     output_tokens = 0
+    stop_reason: Optional[str] = None
 
     async for event in raw_stream:
       if event.type == "message_start":
@@ -603,6 +605,7 @@ async def _generate_content_streaming(
 
       elif event.type == "message_delta":
         output_tokens = event.usage.output_tokens
+        stop_reason = event.delta.stop_reason
 
     # Build the final aggregated response with all content.
     all_parts: list[types.Part] = []
@@ -644,6 +647,7 @@ async def _generate_content_streaming(
             candidates_token_count=output_tokens,
             total_token_count=input_tokens + output_tokens,
         ),
+        finish_reason=to_google_genai_finish_reason(stop_reason),
         partial=False,
     )
 

tests/unittests/models/test_anthropic_llm.py

@@ -1905,3 +1905,153 @@ async def test_streaming_redacted_thinking_block_preserved_in_final():
 
   text_part = final.content.parts[1]
   assert text_part.text == "Done."
+
+
+
+
+# --- Tests for finish_reason ---
+
+
+@pytest.mark.parametrize(
+    "stop_reason,expected_finish_reason",
+    [
+        ("end_turn", types.FinishReason.STOP),
+        ("stop_sequence", types.FinishReason.STOP),
+        ("tool_use", types.FinishReason.STOP),
+        ("max_tokens", types.FinishReason.MAX_TOKENS),
+        ("pause_turn", types.FinishReason.STOP),
+        ("refusal", types.FinishReason.SAFETY),
+        (None, types.FinishReason.FINISH_REASON_UNSPECIFIED),
+    ],
+)
+def test_to_google_genai_finish_reason(stop_reason, expected_finish_reason):
+  """All Anthropic stop_reason values map to the correct ADK FinishReason."""
+  from google.adk.models.anthropic_llm import to_google_genai_finish_reason
+
+  assert to_google_genai_finish_reason(stop_reason) == expected_finish_reason
+
+
+@pytest.mark.asyncio
+async def test_non_streaming_sets_finish_reason():
+  """finish_reason is populated on non-streaming LlmResponse."""
+  llm = AnthropicLlm(model="claude-sonnet-4-20250514")
+  mock_message = anthropic_types.Message(
+      id="msg_test",
+      content=[anthropic_types.TextBlock(text="Hi", type="text", citations=None)],
+      model="claude-sonnet-4-20250514",
+      role="assistant",
+      stop_reason="end_turn",
+      stop_sequence=None,
+      type="message",
+      usage=anthropic_types.Usage(
+          input_tokens=5,
+          output_tokens=2,
+          cache_creation_input_tokens=0,
+          cache_read_input_tokens=0,
+          server_tool_use=None,
+          service_tier=None,
+      ),
+  )
+  mock_client = MagicMock()
+  mock_client.messages.create = AsyncMock(return_value=mock_message)
+
+  llm_request = LlmRequest(
+      model="claude-sonnet-4-20250514",
+      contents=[Content(role="user", parts=[Part.from_text(text="Hi")])],
+      config=types.GenerateContentConfig(system_instruction="Test"),
+  )
+
+  with mock.patch.object(llm, "_anthropic_client", mock_client):
+    responses = [
+        r async for r in llm.generate_content_async(llm_request, stream=False)
+    ]
+
+  assert len(responses) == 1
+  assert responses[0].finish_reason == types.FinishReason.STOP
+
+
+@pytest.mark.asyncio
+async def test_non_streaming_finish_reason_max_tokens():
+  """finish_reason MAX_TOKENS is set when stop_reason is max_tokens."""
+  llm = AnthropicLlm(model="claude-sonnet-4-20250514")
+  mock_message = anthropic_types.Message(
+      id="msg_test",
+      content=[anthropic_types.TextBlock(text="Hi", type="text", citations=None)],
+      model="claude-sonnet-4-20250514",
+      role="assistant",
+      stop_reason="max_tokens",
+      stop_sequence=None,
+      type="message",
+      usage=anthropic_types.Usage(
+          input_tokens=5,
+          output_tokens=2,
+          cache_creation_input_tokens=0,
+          cache_read_input_tokens=0,
+          server_tool_use=None,
+          service_tier=None,
+      ),
+  )
+  mock_client = MagicMock()
+  mock_client.messages.create = AsyncMock(return_value=mock_message)
+
+  llm_request = LlmRequest(
+      model="claude-sonnet-4-20250514",
+      contents=[Content(role="user", parts=[Part.from_text(text="Hi")])],
+      config=types.GenerateContentConfig(system_instruction="Test"),
+  )
+
+  with mock.patch.object(llm, "_anthropic_client", mock_client):
+    responses = [
+        r async for r in llm.generate_content_async(llm_request, stream=False)
+    ]
+
+  assert responses[0].finish_reason == types.FinishReason.MAX_TOKENS
+
+
+@pytest.mark.asyncio
+async def test_streaming_sets_finish_reason():
+  """finish_reason is populated on the final streaming LlmResponse."""
+  llm = AnthropicLlm(model="claude-sonnet-4-20250514")
+
+  events = [
+      MagicMock(
+          type="message_start",
+          message=MagicMock(usage=MagicMock(input_tokens=5, output_tokens=0)),
+      ),
+      MagicMock(
+          type="content_block_start",
+          index=0,
+          content_block=anthropic_types.TextBlock(text="", type="text"),
+      ),
+      MagicMock(
+          type="content_block_delta",
+          index=0,
+          delta=anthropic_types.TextDelta(text="Hi", type="text_delta"),
+      ),
+      MagicMock(type="content_block_stop", index=0),
+      MagicMock(
+          type="message_delta",
+          delta=MagicMock(stop_reason="max_tokens"),
+          usage=MagicMock(output_tokens=1),
+      ),
+      MagicMock(type="message_stop"),
+  ]
+
+  mock_client = MagicMock()
+  mock_client.messages.create = AsyncMock(
+      return_value=_make_mock_stream_events(events)
+  )
+
+  llm_request = LlmRequest(
+      model="claude-sonnet-4-20250514",
+      contents=[Content(role="user", parts=[Part.from_text(text="Hi")])],
+      config=types.GenerateContentConfig(system_instruction="Test"),
+  )
+
+  with mock.patch.object(llm, "_anthropic_client", mock_client):
+    responses = [
+        r async for r in llm.generate_content_async(llm_request, stream=True)
+    ]
+
+  final = responses[-1]
+  assert final.finish_reason == types.FinishReason.MAX_TOKENS