MCP-Mem0: Your Gateway to Long-Term Agent Memory 🚀
@yellnuts
MCP server for long term agent memory with Mem0. Also useful as a template to get you started building your own MCP server with Python!
Overview
What is MCP-Mem0?
MCP-Mem0 is a Python-based server for managing long-term agent memory using Mem0. It also serves as a template for building custom MCP servers. It targets developers who need persistent memory storage for AI agents.
How to use MCP-Mem0?
Clone the repository, install dependencies with pip install -r requirements.txt, and run python server.py. The server starts on http://localhost:5000 and exposes HTTP endpoints for creating, retrieving, and deleting memories. Configuration can be adjusted in config.json.
Key features of MCP-Mem0
- Long-term memory storage and retrieval for agents
- Python‑based, easy to customize and extend
- Lightweight with minimal resource requirements
- Serves as a template for building new MCP servers
- Configurable memory expiry, logging, and port
Use cases of MCP-Mem0
- Persist conversation history for conversational AI agents
- Maintain context across sessions in multi‑turn agent applications
- Serve as a foundation for prototyping custom MCP servers
- Experiment with memory management in research projects
FAQ from MCP-Mem0
What does MCP-Mem0 do?
It provides a server that stores and retrieves long‑term memories for agents via HTTP requests, using the Mem0 system.
What are the system requirements?
Python 3.6 or higher and the packages listed in requirements.txt. No other runtime dependencies are specified.
How do I start the server?
After installing dependencies, run python server.py from the repository root. The server will listen on port 5000 by default.
What API endpoints are available?
Three main endpoints: POST /memory to create a memory, GET /memory/{agent_id} to retrieve it, and DELETE /memory/{agent_id} to delete it.
Can I configure memory expiry?
Yes. Edit the config.json file in the root directory to set parameters like memory_expiry, logging_level, and port.
