nat.plugins.memmachine.memory#

Attributes#

logger

Classes#

MemMachineMemoryClientConfig

Configuration for MemMachine memory client.

Functions#

memmachine_memory_client(...)

Module Contents#

logger#

class MemMachineMemoryClientConfig#

Bases: nat.data_models.memory.MemoryBaseConfig, nat.data_models.retry_mixin.RetryMixin

Configuration for MemMachine memory client.

Based on the MemMachine Python SDK as documented at: MemMachine/MemMachine

Note: This integration is for local/self-hosted MemMachine instances. LLM API keys (e.g., OpenAI) are configured in the MemMachine cfg.yml file, not in this client configuration.

base_url: str#

org_id: str | None = None#

project_id: str | None = None#

timeout: int = 30#

max_retries: int = 3#

async memmachine_memory_client( config: MemMachineMemoryClientConfig, _builder: nat.builder.builder.Builder, ) → collections.abc.AsyncGenerator[nat.memory.interfaces.MemoryEditor, None]#