MemU

You can use the memU MCP Server to access memU AI memory framework features through the MCP protocol. It enables memory storage, smart retrieval, memory management, usage statistics, and multi-user support for AI applications, all via a standardized interface.

How to use

You will interact with the memU MCP Server through an MCP client. The server exposes endpoints for storing and organizing memories, retrieving relevant memories with semantic search, updating and deleting memory data, and obtaining memory usage statistics. It supports multiple users and AI agents, so different workstreams can access shared or separate memory spaces as needed.

How to install

Prerequisites you need before starting are Python 3.8 or newer and a MEMU API key.

Step 1: Clone the server repository and navigate into the project directory.

# Clone the repository
git clone <repository-url>
cd memu_mcp_server

Step 2: Install required Python dependencies.

pip install -r requirements.txt

Step 3: Set your MEMU API key in the environment.

export MEMU_API_KEY="your-memu-api-key"

Step 4: Run the server locally.

python -m memu_mcp_server.main

Step 5: Optional – render deployment. You can deploy to Render using the included blueprint and set MEMU_API_KEY as a secret in the Render dashboard, or deploy via the Render CLI.

# Deploy to Render (via Blueprint)
# Connect your GitHub repository to Render
# Render will detect the render.yaml
# Set MEMU_API_KEY as a secret in Render dashboard
# Deploy!

# Or use the Render CLI
render deploy

Step 6: Other usage variants you may run locally or in testing environments include starting with a specific log level, using a custom config file, or exposing the API for health checks.

Additional deployment and configuration

If you want to run in Render mode for testing locally, you can start the server with a render flag to simulate the cloud environment.

python -m memu_mcp_server.main --log-level DEBUG

You can also run with a custom configuration file if you have specific settings for the server.

python -m memu_mcp_server.main --config config/server.json

Health monitoring and endpoints

When the Web Service component is included, you can access health and status endpoints to monitor health, status, metrics, and service information.

GET /health
GET /status
GET /metrics
GET /info

Configuration and environment

Key configuration involves providing the MEMU API key and choosing how you want to run the server (local development, Render deployment, or containerized deployment). The primary environment variable shown is MEMU_API_KEY.

export MEMU_API_KEY="your-memu-api-key"

Available tools

memorize_conversation

Store conversation memories in memory storage for later retrieval.

retrieve_memory

Retrieve relevant memories using semantic search to support context-aware responses.

search_memory

Search memories by query to find specific conversations or data.

manage_memory

Update or delete memories to keep memory data current and organized.

get_memory_stats

Obtain statistics about memory usage and performance.