ai-memory

Project Overview

ai-memory is a long-term memory solution for AI coding agents. It allows you to seamlessly switch between different agents (such as Claude Code, OpenAI Codex, and Cursor) without re-explaining the architecture, failed approaches, or open questions.

Key Features

• Follows you across agents: 20+ AI coding agents share one memory with real task handoffs
• Follows you across machines: Memory lives in a self-hosted server for seamless desktop-to-laptop transitions
• Works for teams: Multi-user support with shared knowledge per project
• Plain Markdown storage: The source of truth is a git-backed wiki of ordinary .md files
• Silent capture: Lifecycle hooks automatically record prompts, tool calls, and more
• Zero LLM calls: Capture, search, and handoffs work with no API key

Target Audience

• Developers using multiple AI coding agents
• Teams who need to switch work across multiple devices
• Organizations who want to share AI work knowledge across teams

Quick Start

1. Deploy the ai-memory server (local, homelab, or cloud)
2. Configure your AI coding agents to connect to the server
3. Agents automatically capture session information as Markdown pages
4. Continue work on any agent or any device

Core Advantages

• Single self-contained binary
• Works across 20+ AI coding agents
• Plain Markdown storage with no vendor lock-in
• Works with zero LLM calls
• Built-in audit log

System Requirements

• OS: Cross-platform (single binary)