Memu
Memory for 24/7 proactive agents like OpenClaw.
What is Memu?
Memu is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to memory for 24/7 proactive agents like openclaw.
Memory for 24/7 proactive agents like OpenClaw.
This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Memory for 24/7 proactive agents like OpenClaw.
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx memuConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Memu
memU is a persistent memory MCP server designed for 24/7 proactive AI agents such as OpenClaw, providing structured long-term memory that survives across sessions and conversations. It ingests raw data from conversations, documents, images, audio, video, URLs, and logs, then extracts typed memory items — profile facts, events, knowledge, behaviors, skills, and tool usage — that agents can later retrieve with semantic or reasoning-based queries. Teams building always-on AI agents that need to remember user preferences, past decisions, and accumulated knowledge across days or weeks of operation will find memU provides the persistence layer their agents require.
Prerequisites
- Python 3.13 or higher
- pip for installing Python dependencies
- An OpenAI API key (OPENAI_API_KEY) for LLM-based memory extraction and retrieval
- PostgreSQL (optional, for durable persistent storage; defaults to in-memory storage)
- An MCP-compatible agent framework or client such as OpenClaw or Claude Desktop
Clone the memU repository
Clone the repository from GitHub to get the source code and example scripts.
git clone https://github.com/NevaMind-AI/memU.git
cd memUInstall dependencies
Install the Python package in editable mode along with all required dependencies.
pip install -e .Set your OpenAI API key
Export the OPENAI_API_KEY environment variable, which memU uses for memory extraction and LLM-based retrieval. Other OpenAI-compatible providers (OpenRouter, Aliyun DashScope) can also be configured via llm_profiles.
export OPENAI_API_KEY="sk-your-openai-api-key"Configure a custom LLM provider (optional)
If you want to use a non-OpenAI provider, pass llm_profiles when instantiating the MemUService, specifying the base_url, api_key, and model name.
Configure PostgreSQL for durable storage (optional)
By default memU uses in-memory storage, which is lost when the process restarts. For production agents that need true persistence, configure a PostgreSQL database connection.
Add memU to your MCP client configuration
Register memU as an MCP server in your agent's configuration so that the memorize, retrieve, and categories tools are available for the agent to call.
Run the example to verify
Run one of the included example scripts to confirm memU is working correctly before integrating it into your agent.
python examples/example_1_conversation_memory.pyMemu Examples
Client configuration
Add this to your claude_desktop_config.json or agent configuration to register memU as an MCP server. The OPENAI_API_KEY is required.
{
"mcpServers": {
"memu": {
"command": "npx",
"args": ["memu"],
"env": {
"OPENAI_API_KEY": "sk-your-openai-api-key"
}
}
}
}Prompts to try
Use these prompts with an agent that has memU configured to test memory storage and retrieval.
- "Remember that the user prefers dark mode and uses a standing desk at home"
- "What do you know about this user's preferences and past decisions?"
- "Memorize this conversation summary for future reference: [summary]"
- "What skills or tools has this agent used in previous sessions?"
- "Retrieve all events the agent has recorded for user_123 from the past week"Troubleshooting Memu
memorize() or retrieve() fails with OpenAI API errors
Verify OPENAI_API_KEY is set and valid. Check that the key has sufficient quota and that the configured model (default: gpt-4o or similar) is accessible on your account. Use a custom llm_profiles config to switch to a different provider if needed.
Memory is lost when the server restarts
memU defaults to in-memory storage. Configure a PostgreSQL database via the database_config parameter in MemUService to persist memory across restarts. Without PostgreSQL, all stored memories are ephemeral.
Python 3.13 not available on the system
memU requires Python 3.13 or higher. Install it via pyenv (`pyenv install 3.13.0 && pyenv local 3.13.0`) or download it from python.org. Earlier Python versions are not supported.
Frequently Asked Questions about Memu
What is Memu?
Memu is a Model Context Protocol (MCP) server that memory for 24/7 proactive agents like openclaw. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Memu?
Follow the installation instructions on the Memu GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Memu?
Memu works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Memu free to use?
Yes, Memu is open source and available under the NOASSERTION license. You can use it freely in both personal and commercial projects.
Memu Alternatives — Similar Knowledge & Memory Servers
Looking for alternatives to Memu? Here are other popular knowledge & memory servers you can use with Claude, Cursor, and VS Code.
MemPalace
★ 52.6kA local AI memory system that stores all conversations verbatim and organizes them into navigable structures. It provides 19 MCP tools for AI assistants to search and retrieve past decisions, debugging sessions, and architecture debates automatically
Kratos
★ 25.7k🏛️ Memory System for AI Coding Tools - Never explain your codebase again. MCP server with perfect project isolation, 95.8% context accuracy, and the Four Pillars Framework.
Context Mode
★ 15.4kAn MCP server that preserves LLM context by intercepting large data outputs and returning only concise summaries or relevant sections. It enables efficient sandboxed code execution, file processing, and documentation indexing across multiple programm
MemOS
★ 9.3kMemOS (Memory Operating System) is a memory management operating system designed for AI applications. Its goal is: to enable your AI system to have long-term memory like a human, not only remembering what users have said but also actively invoking, u
Everos
★ 5.4kBuild, evaluate, and integrate long-term memory for self-evolving agents.
Cipher
★ 4.8kCipher is an opensource memory layer specifically designed for coding agents. Compatible with Cursor, Windsurf, Claude Desktop, Claude Code, Gemini CLI, AWS's Kiro, VS Code, and Roo Code through MCP, and coding agents, such as Kimi K2.
Browse More Knowledge & Memory MCP Servers
Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Memu in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Memu?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.