strands.memory.extraction.types

Primitive types for the memory extraction subsystem.

ExtractionResult

@dataclass
class ExtractionResult()

Defined in: src/strands/memory/extraction/types.py:39

A discrete entry produced by an :class:Extractor, ready to write via add.

ExtractorContext

@dataclass
class ExtractorContext()

Defined in: src/strands/memory/extraction/types.py:47

Context passed to :meth:Extractor.extract.

Attributes:

• default_model - The agent’s model, supplied so an extractor can default to it.

Extractor

class Extractor(Protocol)

Defined in: src/strands/memory/extraction/types.py:58

Transforms conversation messages into discrete, searchable entries.

Optional on a store’s :class:ExtractionConfig: when absent, the manager passes messages straight to the store’s add_messages (the no-extractor passthrough), which is the right path for backends that extract server-side.

extract

async def extract(
        messages: list[Message],
        context: ExtractorContext | None = None) -> list[ExtractionResult]

Defined in: src/strands/memory/extraction/types.py:66

Extract entries from a batch of messages.

MemoryMessageFilter

@dataclass
class MemoryMessageFilter()

Defined in: src/strands/memory/extraction/types.py:72

Filters content blocks out of messages before extraction.

Blocks whose kind is in :attr:exclude are stripped; a message left with no content is dropped. Defaults to excluding tool traffic (toolUse / toolResult).

ExtractionTriggerContext

@dataclass
class ExtractionTriggerContext()

Defined in: src/strands/memory/extraction/types.py:88

Context handed to :meth:ExtractionTrigger.attach.

Attributes:

• agent - The agent the trigger attaches its hooks to.
• fire - Save this store’s unsaved messages now. Runs in the background and returns immediately. To await completion, see MemoryManager.flush.

ExtractionTrigger

class ExtractionTrigger(ABC)

Defined in: src/strands/memory/extraction/types.py:101

Controls when a store’s :class:ExtractionConfig runs.

A trigger is a self-attaching value object: :meth:attach wires the agent hooks it needs and calls :attr:ExtractionTriggerContext.fire when extraction should happen. Subclass for custom triggering logic. A trigger that never fires never extracts; for a guaranteed final write, use MemoryManager.flush.

Attributes:

• name - Stable identifier for this trigger kind, used in logging.

attach

@abstractmethod
def attach(context: ExtractionTriggerContext) -> None

Defined in: src/strands/memory/extraction/types.py:117

Wire this trigger into the agent lifecycle.

Called once per store during MemoryManager initialization. Register hooks on context.agent and call context.fire() when extraction should run.

ExtractionConfig

class ExtractionConfig(TypedDict)

Defined in: src/strands/memory/extraction/types.py:127

Per-store automatic-extraction configuration.

Attributes:

• trigger - When to run extraction. A single trigger or a list; multiple triggers compose (extraction runs whenever any fires). Omit to default to every 5 turns; an explicit empty list is rejected at construction.
• extractor - How to turn messages into entries. When set, the store must implement add. When omitted, the default depends on the store’s write methods: a store implementing only add defaults to a :class:~strands.memory.extraction.model_extractor.ModelExtractor that distills facts client-side, while a store implementing add_messages uses server-side extraction (the manager hands the filtered messages straight to add_messages, no model call).
• filter - Content blocks to strip before extraction. Defaults to :data:DEFAULT_MEMORY_MESSAGE_FILTER (excludes toolUse / toolResult). Pass MemoryMessageFilter(exclude=[]) to keep tool blocks.