memmy-agent
🍙 A personal AI agent & local memory hub for all AI agents, gives every AI one shared, fully controlled memory and persistent context — all AI remember the sa
What is memmy-agent?
🍙 A personal AI agent & local memory hub for all AI agents, gives every AI one shared, fully controlled memory and persistent context — all AI remember the same you. Now supports Claude Code, Codex, OpenClaw and Hermes Agent etc.
Overview of memmy-agent
What it is
Memmy is a personal AI agent and local memory hub designed to provide a shared, controlled memory and persistent context for multiple AI agents. It enables cross-agent task continuity, allowing different AI tools to remember the same user and maintain consistent context across sessions. The project is open-source under the MIT license and supports integration with agents like Claude Code, Codex, OpenClaw, Hermes Agent, and others.
Key features
- Cross-agent task continuity: Continue the same work across different AI agents such as DeepSeek Harness, Claude Code, Codex, Cursor, and more.
- Local memory service: A private memory hub that runs locally, giving users full control over their data.
- Multiple interfaces: Available as a desktop app, CLI/TUI, and a dedicated
memmy-memoryCLI for scripting and debugging. - OpenAI-compatible API: Exposes an API endpoint at port 18990 for integration with other tools.
- BYOK (Bring Your Own Key) support: Users can configure their own model providers and API keys.
- Systemd service management: On Linux, the installer sets up user-level systemd services for persistent operation.
Use cases
- Developers who want a unified memory across multiple AI coding assistants.
- Users who need persistent context for AI agents across different sessions and tools.
- Teams planning to share knowledge between AI assistants under privacy protection.
- Anyone seeking a local, self-hosted memory solution for AI agents.
Getting started
For complete installation and configuration instructions, see the Getting Started guide in the repository.On Linux x64 or arm64 with Node.js 22 or newer and an available systemd user session:bashcurl -fsSL https://raw.githubusercontent.com/MemTensor/memmy-agent/main/scripts/install.sh | bashmemmyTo start from source code:bashgit clone https://github.com/MemTensor/memmy-agent.gitcd memmy-agentcp .env.example .envnpm installnpm run buildbash scripts/dev-start.sh
License and pricing
The project is licensed under the MIT License. The README mentions trial credits for the desktop app and a BYOK mode for using your own model API, but no specific pricing is stated.
Based on the project's public README and repository metadata.