fleet-memory/memora-clients/python

memora-client

Python client for Memora - Semantic memory system with personality-driven thinking.

Auto-generated from OpenAPI spec - provides type-safe access to all Memora API endpoints.

Installation

pip install memora-client

Quick Start

from agent_memory_api_client import Client
from agent_memory_api_client.api.memory_storage import put_api_put_post
from agent_memory_api_client.api.reasoning import think_api_think_post

client = Client(base_url="http://localhost:8000")

# Store memory
put_api_put_post.sync(
    client=client,
    body={
        "agent_id": "user123",
        "content": "Alice loves machine learning"
    }
)

# Think (generate answer with personality)
response = think_api_think_post.sync(
    client=client,
    body={
        "agent_id": "user123",
        "query": "What does Alice think about AI?",
        "thinking_budget": 50
    }
)
print(response.text)

Async Support

from agent_memory_api_client import Client
from agent_memory_api_client.api.reasoning import think_api_think_post

async with Client(base_url="http://localhost:8000") as client:
    response = await think_api_think_post.asyncio(
        client=client,
        body={
            "agent_id": "user123",
            "query": "What does Alice think about AI?"
        }
    )
    print(response.text)

API Modules

This client provides access to:

• memory_storage - Store and retrieve facts
• search - Semantic and temporal search
• reasoning - Personality-driven thinking
• visualization - Memory graphs and statistics
• management - Agent profiles and configuration
• documents - Document tracking

See auto-generated code for full API surface and type hints.

Development

Auto-generated from openapi.json. See RELEASE.md for regeneration instructions.

Links

• GitHub Repository
• Full Documentation