Memorymesh

Memorymesh is an open-source Python library that gives any LLM (large language model) persistent, structured memory across sessions, tools, and projects, with zero external dependencies and a privacy-first approach. It stores relevant facts, user preferences, and decisions locally using SQLite, and provides a simple API (remember(), recall(), forget()) for developers to seamlessly embed memory in AI applications. With integration via MCP server, it syncs across multiple AI tools (Claude, Gemini, Cursor, etc.), and supports optional encrypted storage, pluggable embeddings (local, Ollama, OpenAI), and runs fully locally for maximum privacy. All data remains on the user's machine, and the library works with any LLM framework or API.