memU MCP Server

A Model Context Protocol (MCP) server that provides access to memU AI memory framework capabilities.

Overview

This MCP server wraps the memU AI memory framework, enabling AI applications to use advanced memory management features through the standardized MCP protocol.

Features

Memory Storage: Store and organize conversation memories
Smart Retrieval: Retrieve relevant memories using semantic search
Memory Management: Update, delete, and organize memory data
Statistics: Get insights into memory usage and performance
Multi-user Support: Handle multiple users and AI agents

Quick Start

Prerequisites

Python 3.8+
memU API key (get one at https://app.memu.so/api-key/)

Local Development

# Clone the repository
git clone <repository-url>
cd memu-mcp-server

# Install dependencies
pip install -r requirements.txt

# Set up environment variables
export MEMU_API_KEY="your-memu-api-key"

# Run the server
python -m memu_mcp_server.main

Render Deployment

# Deploy to Render (using Blueprint)
1. Connect your GitHub repository to Render
2. Render will automatically detect render.yaml
3. Set MEMU_API_KEY as a secret in Render dashboard
4. Deploy!

# Or use the Render CLI
render deploy

Usage Examples

# Local development
python -m memu_mcp_server.main --log-level DEBUG

# Render mode (for testing locally)
python -m memu_mcp_server.main --render-mode

# With custom configuration
python -m memu_mcp_server.main --config config/server.json

# API server (for health checks)
python -m memu_mcp_server.api --host 0.0.0.0 --port 8080

Configuration

Local Development: See config/example.json for configuration options
Render Deployment: See Render Deployment Guide
Environment Variables: See Environment Variables Guide

Available Tools

memorize_conversation: Store conversation memories
retrieve_memory: Retrieve relevant memories
search_memory: Search memories by query
manage_memory: Update or delete memories
get_memory_stats: Get memory statistics

Documentation

API Reference - Detailed API documentation
Setup Guide - Installation and configuration
Render Deployment - Deploy to Render platform
Environment Variables - Configuration reference

Deployment Options

Local Development

python -m memu_mcp_server.main

Docker

docker-compose up memu-mcp-server

Render (Cloud)

Use the included render.yaml Blueprint for one-click deployment to Render.

Claude Desktop Integration

Add to your Claude Desktop configuration:

{
  "mcpServers": {
    "memu-memory": {
      "command": "python",
      "args": ["-m", "memu_mcp_server.main"],
      "env": {
        "MEMU_API_KEY": "your_api_key_here"
      }
    }
  }
}

Health Monitoring

When deployed with the Web Service component, monitoring endpoints are available:

GET /health - Health check
GET /status - Detailed status
GET /metrics - Performance metrics
GET /info - Service information

Contributing

Fork the repository
Create a feature branch
Make your changes
Add tests if applicable
Submit a pull request

Support

GitHub Issues: Report bugs and feature requests
Documentation: Check the docs/ directory
Email: support@example.com

License

MIT License

Quick Start

Prerequisites

Local Development

# Clone the repository git clone <repository-url> cd memu-mcp-server # Install dependencies pip install -r requirements.txt # Set up environment variables export MEMU_API_KEY="your-memu-api-key" # Run the server python -m memu_mcp_server.main

Render Deployment

# Deploy to Render (using Blueprint) 1. Connect your GitHub repository to Render 2. Render will automatically detect render.yaml 3. Set MEMU_API_KEY as a secret in Render dashboard 4. Deploy! # Or use the Render CLI render deploy

Usage Examples

# Local development python -m memu_mcp_server.main --log-level DEBUG # Render mode (for testing locally) python -m memu_mcp_server.main --render-mode # With custom configuration python -m memu_mcp_server.main --config config/server.json # API server (for health checks) python -m memu_mcp_server.api --host 0.0.0.0 --port 8080

Deployment Options

Local Development

python -m memu_mcp_server.main

Docker

docker-compose up memu-mcp-server

Render (Cloud)

Use the included render.yaml Blueprint for one-click deployment to Render.

Claude Desktop Integration

Add to your Claude Desktop configuration:

{ "mcpServers": { "memu-memory": { "command": "python", "args": ["-m", "memu_mcp_server.main"], "env": { "MEMU_API_KEY": "your_api_key_here" } } } }