Memory operating system for LLMs and AI agents that stores, retrieves, and manages long-term memory as an editable graph.
MemTensor/memmy-agent
A local, self-hosted memory hub that gives multiple AI agents such as Claude Code, Codex and OpenClaw one shared, persistent memory of you.
- Stars
- 2,131
- Forks
- 210
- Open issues
- 62
- Last commit
- 3 hours ago
What is it?
Memmy is a personal AI agent and local memory hub. It records your AI collaboration history and turns it into structured memory, so different agents can pick up the same project background, preferences and unfinished progress instead of starting from zero each time.
The project connects to a range of agents, including Claude Code, Codex, Cursor, OpenClaw and Hermes Agent. It is offered as a desktop app, a `memmy` command-line and TUI client, and a separate `memmy-memory` CLI that talks to a local memory service on localhost. A `memmy serve` command exposes an OpenAI-compatible API on port 18990.
Memmy can run in bring-your-own-key mode, where you supply your own model provider credentials in a YAML config file. The project is written in TypeScript, distributed under the MIT license, and targets Linux, with source builds requiring Node.js 22 or newer.
Key features
- 01Shares one persistent memory across Claude Code, Codex, Cursor, OpenClaw and other agents.
- 02Turns local AI collaboration history into structured, searchable memory.
- 03Stores memory in a local service bound to localhost, with user-level systemd services on Linux.
- 04Provides a desktop app, a `memmy` CLI with TUI, and a `memmy-memory` CLI.
- 05Exposes an OpenAI-compatible API through `memmy serve`.
- 06Supports bring-your-own-key model configuration in a YAML file.
Use cases
- Switching between coding agents without re-explaining your project each time.
- Keeping preferences and progress consistent across several AI tools on one machine.
- Giving scripts or local agents a shared memory store through the memory CLI.
- Running an OpenAI-compatible endpoint backed by your own model keys.
Topics
FAQ
- What is memmy-agent?
- A local, self-hosted memory hub that gives multiple AI agents such as Claude Code, Codex and OpenClaw one shared, persistent memory of you. Memmy is a personal AI agent and local memory hub. It records your AI collaboration history and turns it into structured memory, so different agents can pick up the same project background, preferences and unfinished progress instead of starting from zero each time.
- Is memmy-agent free to use? What license does it have?
- memmy-agent is open source under the MIT license. Check the license terms before using it in commercial products.
- Is memmy-agent still maintained?
- Actively maintained. The last commit to memmy-agent was on Oct 22, 2026.
- What can I use memmy-agent for?
- Switching between coding agents without re-explaining your project each time. Keeping preferences and progress consistent across several AI tools on one machine. Giving scripts or local agents a shared memory store through the memory CLI. Running an OpenAI-compatible endpoint backed by your own model keys.
- Where can I get memmy-agent?
- The source code and installation instructions are on GitHub: https://github.com/MemTensor/memmy-agent
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
An open-source long-term memory layer for AI agents that stores and recalls information across sessions via a Python SDK, REST API, and MCP server.
A Python-native agent harness and terminal coding assistant that works with many cloud and local LLMs, with tools, memory, skills and a Web UI.
A personal AI agent that builds a correctable model of its user over time and helps shape long-running life and work goals into Paths.
Self-hosted private 1:1 chat channel linking a phone PWA to your AI companion, running via Claude Code or any OpenAI-compatible LLM.
Open-source, inspectable framework for building an AI Agent Twin that distills your knowledge, memories and decisions into a grounded personal assistant.