client.memory

client.memory — the memory operations.

On AsyncKindgi every method is the same, awaited.

client.memory.list_facts()

list_facts(
    *,
    limit: int | None = None,
    cursor: str | None = None,
    type: str | None = None,
    scope: str | None = None,
    scope_kind: Literal['tenant', 'org', 'project'] | None = None,
    scope_id: str | None = None,
    inherit: bool | None = None,
    timeout: float | None = None,
) -> FactCollectionPage

List facts. GET /v1/memory/facts

Cursor-paginated. Filters: ?type= (exact match on Fact.type), ?scope= (URL-encoded JSON partial memory Scope), ?scopeKind + ?scopeId + ?inherit (discriminated PlatformScope triplet — threaded into the binding as platformScope; both fields coexist). Sort order is binding-defined.

client.memory.write_fact()

write_fact(
    *,
    idempotency_key: str | None = None,
    timeout: float | None = None,
    **fields: Any,
) -> Fact

Write a fact. POST /v1/memory/facts

Persists a new fact. type selects the retrieval policy (which indexes populate). Semantic-indexed types require an embedding provider bound on the deployment; if unavailable, the route returns 400 bad-input.

client.memory.get_fact()

get_fact(*, timeout: float | None = None) -> Fact

Fetch a fact. GET /v1/memory/facts/{factId}

client.memory.supersede_fact()

supersede_fact(
    *,
    idempotency_key: str | None = None,
    timeout: float | None = None,
) -> SupersedeFactResult

Mark a fact as superseded. POST /v1/memory/facts/{factId}/supersede

Soft-delete via supersession — the historical row is retained until retention sweeps remove it. Idempotent: superseding an already-superseded fact returns 200 { superseded: true }.

client.memory.retrieve()

retrieve(
    *,
    idempotency_key: str | None = None,
    timeout: float | None = None,
    **fields: Any,
) -> RetrieveMemoryResult

Retrieve facts by intent. POST /v1/memory/retrieve

Cross-history retrieval. Body is a RetrieveIntent shape mirroring the agent-side declarative retrieval. Semantic modes require an embedding provider bound on the deployment.