fix: handle mcp tool calls in previous response correctly (#3155)

# What does this PR do?

Handles MCP tool calls in a previous response

Closes #3105

## Test Plan
Made call to create response with tool call, then made second call with
the first linked through previous_response_id. Did not get error.

Also added unit test.

Signed-off-by: Gordon Sim <gsim@redhat.com>

llama_stack/providers/inline/agents/meta_reference/responses/utils.py

@@ -17,6 +17,8 @@ from llama_stack.apis.agents.openai_responses import (
     OpenAIResponseOutputMessageContent,
     OpenAIResponseOutputMessageContentOutputText,
     OpenAIResponseOutputMessageFunctionToolCall,
+    OpenAIResponseOutputMessageMCPCall,
+    OpenAIResponseOutputMessageMCPListTools,
     OpenAIResponseText,
 )
 from llama_stack.apis.inference import (
@@ -117,6 +119,25 @@ async def convert_response_input_to_chat_messages(
                     ),
                 )
                 messages.append(OpenAIAssistantMessageParam(tool_calls=[tool_call]))
+            elif isinstance(input_item, OpenAIResponseOutputMessageMCPCall):
+                tool_call = OpenAIChatCompletionToolCall(
+                    index=0,
+                    id=input_item.id,
+                    function=OpenAIChatCompletionToolCallFunction(
+                        name=input_item.name,
+                        arguments=input_item.arguments,
+                    ),
+                )
+                messages.append(OpenAIAssistantMessageParam(tool_calls=[tool_call]))
+                messages.append(
+                    OpenAIToolMessageParam(
+                        content=input_item.output,
+                        tool_call_id=input_item.id,
+                    )
+                )
+            elif isinstance(input_item, OpenAIResponseOutputMessageMCPListTools):
+                # the tool list will be handled separately
+                pass
             else:
                 content = await convert_response_content_to_chat_content(input_item.content)
                 message_type = await get_message_type_by_role(input_item.role)

tests/integration/non_ci/responses/test_basic_responses.py

@@ -7,8 +7,9 @@
 import time
 
 import pytest
-from fixtures.test_cases import basic_test_cases, image_test_cases, multi_turn_image_test_cases, multi_turn_test_cases
+
-from streaming_assertions import StreamingValidator
+from .fixtures.test_cases import basic_test_cases, image_test_cases, multi_turn_image_test_cases, multi_turn_test_cases
+from .streaming_assertions import StreamingValidator
 
 
 @pytest.mark.parametrize("case", basic_test_cases)

tests/integration/non_ci/responses/test_tool_responses.py

@@ -10,7 +10,12 @@ import os
 import httpx
 import openai
 import pytest
-from fixtures.test_cases import (
+
+from llama_stack import LlamaStackAsLibraryClient
+from llama_stack.core.datatypes import AuthenticationRequiredError
+from tests.common.mcp import dependency_tools, make_mcp_server
+
+from .fixtures.test_cases import (
     custom_tool_test_cases,
     file_search_test_cases,
     mcp_tool_test_cases,
@@ -18,12 +23,8 @@ from fixtures.test_cases import (
     multi_turn_tool_execution_test_cases,
     web_search_test_cases,
 )
-from helpers import new_vector_store, setup_mcp_tools, upload_file, wait_for_file_attachment
+from .helpers import new_vector_store, setup_mcp_tools, upload_file, wait_for_file_attachment
-from streaming_assertions import StreamingValidator
+from .streaming_assertions import StreamingValidator
-
-from llama_stack import LlamaStackAsLibraryClient
-from llama_stack.core.datatypes import AuthenticationRequiredError
-from tests.common.mcp import dependency_tools, make_mcp_server
 
 
 @pytest.mark.parametrize("case", web_search_test_cases)
@@ -195,6 +196,56 @@ def test_response_non_streaming_mcp_tool(compat_client, text_model_id, case):
         assert len(response.output) >= 3
 
 
+@pytest.mark.parametrize("case", mcp_tool_test_cases)
+def test_response_sequential_mcp_tool(compat_client, text_model_id, case):
+    if not isinstance(compat_client, LlamaStackAsLibraryClient):
+        pytest.skip("in-process MCP server is only supported in library client")
+
+    with make_mcp_server() as mcp_server_info:
+        tools = setup_mcp_tools(case.tools, mcp_server_info)
+
+        response = compat_client.responses.create(
+            model=text_model_id,
+            input=case.input,
+            tools=tools,
+            stream=False,
+        )
+
+        assert len(response.output) >= 3
+        list_tools = response.output[0]
+        assert list_tools.type == "mcp_list_tools"
+        assert list_tools.server_label == "localmcp"
+        assert len(list_tools.tools) == 2
+        assert {t.name for t in list_tools.tools} == {
+            "get_boiling_point",
+            "greet_everyone",
+        }
+
+        call = response.output[1]
+        assert call.type == "mcp_call"
+        assert call.name == "get_boiling_point"
+        assert json.loads(call.arguments) == {
+            "liquid_name": "myawesomeliquid",
+            "celsius": True,
+        }
+        assert call.error is None
+        assert "-100" in call.output
+
+        # sometimes the model will call the tool again, so we need to get the last message
+        message = response.output[-1]
+        text_content = message.content[0].text
+        assert "boiling point" in text_content.lower()
+
+        response2 = compat_client.responses.create(
+            model=text_model_id, input=case.input, tools=tools, stream=False, previous_response_id=response.id
+        )
+
+        assert len(response2.output) >= 1
+        message = response2.output[-1]
+        text_content = message.content[0].text
+        assert "boiling point" in text_content.lower()
+
+
 @pytest.mark.parametrize("case", custom_tool_test_cases)
 def test_response_non_streaming_custom_tool(compat_client, text_model_id, case):
     response = compat_client.responses.create(

tests/unit/providers/agents/meta_reference/test_openai_responses.py

@@ -24,6 +24,7 @@ from llama_stack.apis.agents.openai_responses import (
     OpenAIResponseMessage,
     OpenAIResponseObjectWithInput,
     OpenAIResponseOutputMessageContentOutputText,
+    OpenAIResponseOutputMessageMCPCall,
     OpenAIResponseOutputMessageWebSearchToolCall,
     OpenAIResponseText,
     OpenAIResponseTextFormat,
@@ -461,6 +462,53 @@ async def test_prepend_previous_response_web_search(openai_responses_impl, mock_
     assert input[3].content == "fake_input"
 
 
+async def test_prepend_previous_response_mcp_tool_call(openai_responses_impl, mock_responses_store):
+    """Test prepending a previous response which included an mcp tool call to a new response."""
+    input_item_message = OpenAIResponseMessage(
+        id="123",
+        content=[OpenAIResponseInputMessageContentText(text="fake_previous_input")],
+        role="user",
+    )
+    output_tool_call = OpenAIResponseOutputMessageMCPCall(
+        id="ws_123",
+        name="fake-tool",
+        arguments="fake-arguments",
+        server_label="fake-label",
+    )
+    output_message = OpenAIResponseMessage(
+        id="123",
+        content=[OpenAIResponseOutputMessageContentOutputText(text="fake_tool_call_response")],
+        status="completed",
+        role="assistant",
+    )
+    response = OpenAIResponseObjectWithInput(
+        created_at=1,
+        id="resp_123",
+        model="fake_model",
+        output=[output_tool_call, output_message],
+        status="completed",
+        text=OpenAIResponseText(format=OpenAIResponseTextFormat(type="text")),
+        input=[input_item_message],
+    )
+    mock_responses_store.get_response_object.return_value = response
+
+    input_messages = [OpenAIResponseMessage(content="fake_input", role="user")]
+    input = await openai_responses_impl._prepend_previous_response(input_messages, "resp_123")
+
+    assert len(input) == 4
+    # Check for previous input
+    assert isinstance(input[0], OpenAIResponseMessage)
+    assert input[0].content[0].text == "fake_previous_input"
+    # Check for previous output MCP tool call
+    assert isinstance(input[1], OpenAIResponseOutputMessageMCPCall)
+    # Check for previous output web search response
+    assert isinstance(input[2], OpenAIResponseMessage)
+    assert input[2].content[0].text == "fake_tool_call_response"
+    # Check for new input
+    assert isinstance(input[3], OpenAIResponseMessage)
+    assert input[3].content == "fake_input"
+
+
 async def test_create_openai_response_with_instructions(openai_responses_impl, mock_inference_api):
     # Setup
     input_text = "What is the capital of Ireland?"