SelfManagedStrategyProps

class aws_cdk.aws_bedrock_agentcore_alpha.SelfManagedStrategyProps(*, name, description=None, invocation_configuration, historical_context_window_size=None, trigger_conditions=None)

Bases: MemoryStrategyCommonProps

(experimental) Configuration parameters for a self managed memory strategy existing built-in default prompts/models.

Parameters:

  • name (str) – (experimental) The name for the strategy.

  • description (Optional[str]) – (experimental) The description of the strategy. Default: No description

  • invocation_configuration (Union[InvocationConfiguration, Dict[str, Any]]) – (experimental) Invocation configuration for self managed memory strategy.

  • historical_context_window_size (Union[int, float, None]) – (experimental) Define the number of previous events to be included when processing memory. A larger history window provides more context from past conversations. Default: - 4

  • trigger_conditions (Union[TriggerConditions, Dict[str, Any], None]) – (experimental) Trigger conditions for self managed memory strategy. Default: - undefined

Stability:

experimental

ExampleMetadata:

fixture=default infused

Example:

bucket = s3.Bucket(self, "memoryBucket",
    bucket_name="test-memory",
    removal_policy=cdk.RemovalPolicy.DESTROY,
    auto_delete_objects=True
)

topic = sns.Topic(self, "topic")

# Create a custom semantic memory strategy
self_managed_strategy = agentcore.MemoryStrategy.using_self_managed(
    name="selfManagedStrategy",
    description="self managed memory strategy",
    historical_context_window_size=5,
    invocation_configuration=agentcore.InvocationConfiguration(
        topic=topic,
        s3_location=s3.Location(
            bucket_name=bucket.bucket_name,
            object_key="memory/"
        )
    ),
    trigger_conditions=agentcore.TriggerConditions(
        message_based_trigger=1,
        time_based_trigger=cdk.Duration.seconds(10),
        token_based_trigger=100
    )
)

# Create memory with custom strategy
memory = agentcore.Memory(self, "MyMemory",
    memory_name="my-custom-memory",
    description="Memory with custom strategy",
    expiration_duration=cdk.Duration.days(90),
    memory_strategies=[self_managed_strategy]
)

Attributes

description

(experimental) The description of the strategy.

Default:

No description

Stability:

experimental

Required:

  • No

historical_context_window_size

(experimental) Define the number of previous events to be included when processing memory.

A larger history window provides more context from past conversations.

Default:

  • 4

Stability:

experimental

Required:

  • No

invocation_configuration

(experimental) Invocation configuration for self managed memory strategy.

Stability:

experimental

Required:

  • Yes

name

(experimental) The name for the strategy.

Stability:

experimental

Required:

  • Yes

trigger_conditions

(experimental) Trigger conditions for self managed memory strategy.

Default:

  • undefined

Stability:

experimental

Required:

  • No