Mnemosyne lets users connect any LLM client to an MCP server. Information shared in conversations is automatically sent to Mnemosyne, chunked, embedded, and stored in tiered memory. When a new conversation starts, relevant memories are retrieved and injected into the system prompt, giving the LLM persistent memory across sessions.

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "mnemosyne": {
      "command": "uvx",
      "args": [
        "mnemosyne"
      ]
    }
  }
}

Or install with pip: pip install mnemosyne

README Excerpt

A persistent memory layer for LLM applications. Stores, retrieves, and injects relevant context from past conversations so your AI assistant remembers who you are and what you care about. Exposes a [Model Context Protocol (MCP)](https://modelcontextprotocol.io) server that any MCP-compatible client (Claude Desktop, Cursor, etc.) can connect to.

Tools (13)

LLM_PROVIDERanthropicepisodicforget_memory_toolgeminiingest_memory_toollocallong_termopenairetrieve_memory_toolsemanticsession_init_toolworking

Mnemosyne

Quick Install

README Excerpt

Tools (13)

Related AI/LLM Servers