memos-sdk 0.1.0

Python SDK for the Distributed Memory OS (MemOS)

MemOS Python SDK 🧠

Official Python client for Distributed MemOS, a production-ready cognitive memory infrastructure for autonomous AI systems.

Installation

pip install memos-sdk

Quick Start

from memos_sdk import MemOSClient, MemoryType

# Connect to the MemOS Cluster
client = MemOSClient("localhost:50051")

tenant_id = "00000000-0000-0000-0000-000000000001"
agent_id = "00000000-0000-0000-0000-000000000002"

# 1. Store a Memory
memory_id = client.store(
    tenant_id=tenant_id,
    agent_id=agent_id,
    content="The user prefers a high-contrast dark mode with large fonts.",
    memory_type=MemoryType.MEMORY_TYPE_EPISODIC,
    importance=0.85
)
print(f"Stored Memory: {memory_id}")

# 2. Retrieve with Cognitive Search
results = client.retrieve(
    tenant_id=tenant_id,
    agent_id=agent_id,
    query="UI preferences dark mode",
    limit=3
)

for res in results:
    print(f"[{res.score:.2f}] {res.memory.content}")

Features

• gRPC based high-performance communication.
• Supports Cognitive Ranking (Semantic + Temporal + Importance).
• Thread-safe client implementation.
• Context manager support (with MemOSClient() as client:).