thedotmack/claude-mem

What it does

claude-mem is a persistent memory layer for AI coding agents: it captures everything an agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. It works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode and more.

What's inside

Implemented in TypeScript, the repo contains modules for session capture, AI compression, memory storage, and context injection, plus adapters and configuration for multiple agents.

Tech stack

Primary language is TypeScript on the Node.js ecosystem; it relies on AI models for summarization and provides integration adapters for different agents.

Use cases

• Developers using Claude Code and similar agents who want to avoid re-explaining context
• Users of AI coding workflows that need project context to persist across sessions
• Engineering teams adding a memory layer to custom agents to cut repeated prompting
• Enthusiasts studying agent memory and context-compression mechanisms

Why it's trending

With 578 stars added today, nearly 98k total, and a #8 spot on GitHub Trending daily, it shows that 'making agents remember context' is a pressing need in today's AI coding toolchain.

Current chart placements