Update

Author: dirkbrnd

libs/agno/tests/integration/models/anthropic/test_tool_use.py

@@ -255,5 +255,5 @@ def test_tool_call_list_parameters():
             tool_calls.extend(msg.tool_calls)
     for call in tool_calls:
         if call.get("type", "") == "function":
-            assert call["function"]["name"] in ["get_contents", "exa_answer", "search_exa"]
+            assert call["function"]["name"] in ["get_contents", "exa_answer", "search_exa", "find_similar"]
     assert response.content is not None

libs/agno/tests/integration/models/openai/responses/test_tool_use.py

@@ -153,22 +153,28 @@ def test_parallel_tool_calls():
 
 def test_multiple_tool_calls():
     """Test multiple different tool types with the responses API."""
+    def get_the_weather(city: str):
+        return f"It is currently 70 degrees and cloudy in {city}"
+
+    def get_favourite_city():
+        return "Tokyo"
+
     agent = Agent(
         model=OpenAIResponses(id="gpt-4.1-mini"),
-        tools=[YFinanceTools(cache_results=True), DuckDuckGoTools(cache_results=True)],
+        tools=[get_the_weather, get_favourite_city],
         markdown=True,
         telemetry=False,
         monitoring=False,
     )
 
-    response = agent.run("Find the latest price of TSLA. Then, based on the price, find the latest news about it.")
+    response = agent.run("Find my favourite city. Then, get the weather in that city.")
 
     # Verify tool usage
     tool_calls = [msg.tool_calls for msg in response.messages if msg.tool_calls]
-    assert len(tool_calls) >= 2  # At least two messages have tool calls
-    assert sum(len(calls) for calls in tool_calls) >= 2  # Total of 2 tool calls made
+    assert len(tool_calls) >= 2  # At least one message has tool calls
+    assert sum(len(calls) for calls in tool_calls) >= 1  # Total of 1 tool calls made
     assert response.content is not None
-    assert "TSLA" in response.content and "latest news" in response.content.lower()
+    assert "Tokyo" in response.content and "70" in response.content
 
 
 def test_tool_call_custom_tool_no_parameters():

libs/agno/tests/integration/models/xai/test_basic.py

@@ -29,17 +29,15 @@ def test_basic():
 
 
 def test_basic_stream():
-    agent = Agent(model=xAI(id="grok-3-mini-fast"), markdown=True, telemetry=False, monitoring=False)
+    agent = Agent(model=xAI(id="grok-3-mini"), markdown=True, telemetry=False, monitoring=False)
 
     response_stream = agent.run("Share a 2 sentence horror story", stream=True)
 
     # Verify it's an iterator
     assert hasattr(response_stream, "__iter__")
 
-    responses = list(response_stream)
-    assert len(responses) > 0
-    for response in responses:
-        assert response.content is not None
+    for response in response_stream:
+        assert response.content is not None or response.reasoning_content is not None
 
     _assert_metrics(agent.run_response)
 
@@ -63,7 +61,7 @@ async def test_async_basic_stream():
     response_stream = await agent.arun("Share a 2 sentence horror story", stream=True)
 
     async for response in response_stream:
-        assert response.content is not None
+        assert response.content is not None or response.reasoning_content is not None
 
     _assert_metrics(agent.run_response)