When will the agent use the contents of the memory?

[zhangjy859]

Hi,

I am trying to create a agent with memory, I use the demo code from document

# ...
# Use a temporary directory for ChromaDB persistence
with tempfile.TemporaryDirectory() as tmpdir:
    chroma_user_memory = ChromaDBVectorMemory(
        config=PersistentChromaDBVectorMemoryConfig(
            collection_name="preferences",
            persistence_path=tmpdir,  # Use the temp directory here
            k=2,  # Return top k results
            score_threshold=0.4,  # Minimum similarity score
            embedding_function_config=SentenceTransformerEmbeddingFunctionConfig(
                model_name="all-MiniLM-L6-v2"  # Use default model for testing
            ),
        )
    )
    # Add user preferences to memory
    asyncio.run(chroma_user_memory.add(
        MemoryContent(
            content="The weather should be in metric units",
            mime_type=MemoryMimeType.TEXT,
            metadata={"category": "preferences", "type": "units"},
        )
    ))

    asyncio.run(chroma_user_memory.add(
        MemoryContent(
            content="Meal recipe must be vegan",
            mime_type=MemoryMimeType.TEXT,
            metadata={"category": "preferences", "type": "dietary"},
        )
    ))


    async def get_weather(city: str, units: str = "imperial") -> str:
        if units == "imperial":
            return f"The weather in {city} is 73 °F and Sunny."
        elif units == "metric":
            return f"The weather in {city} is 23 °C and Sunny."
        else:
            return f"Sorry, I don't know the weather in {city}."

    # Create assistant agent with ChromaDB memory
    assistant_agent = AssistantAgent(
        name="assistant_agent",
        model_client=custom_model_client,
        tools=[get_weather],
        memory=[chroma_user_memory],
    )

    stream = assistant_agent.run_stream(task="What's the weather in New York? ")
    asyncio.run(Console(stream))

    asyncio.run(custom_model_client.close())
    asyncio.run(chroma_user_memory.close())

Its ok, and output

---------- TextMessage (user) ----------
What's the weather in New York? 
---------- MemoryQueryEvent (assistant_agent) ----------
[MemoryContent(content='The weather should be in metric units', mime_type='MemoryMimeType.TEXT', metadata={'mime_type': 'MemoryMimeType.TEXT', 'category': 'preferences', 'type': 'units', 'score': 0.41220688819885254, 'id': '61a4be3c-b2c6-4a03-8e6e-9e31689bbbbe'})]
---------- ToolCallRequestEvent (assistant_agent) ----------
[FunctionCall(id='call_98806457', arguments='{"city":"New York","units":"metric"}', name='get_weather')]
---------- ToolCallExecutionEvent (assistant_agent) ----------
[FunctionExecutionResult(content='The weather in New York is 23 °C and Sunny.', name='get_weather', call_id='call_98806457', is_error=False)]
---------- ToolCallSummaryMessage (assistant_agent) ----------
The weather in New York is 23 °C and Sunny.

But, if we change the task promopt

stream = assistant_agent.run_stream(task="How are you?")

output:

---------- TextMessage (user) ----------
How are you?
---------- ThoughtEvent (assistant_agent) ----------
...
---------- TextMessage (assistant_agent) ----------
I'm doing well, thank you! How can I assist you today?

I noticed that the agent is no longer generating MemoryQueryEvent. Does this mean that the agent will not further use the content in the Memory? But if it doesn't even Query the content in the memory, how can it determine whether to perform a MemoryQuery or not?