MCP.so
ログイン
M

Mem0 General

@ryaker

Mem0 General について

OpenAI just added memory across your chats across your openAI account. But wouldn't it be awesome to have general AI memory across all your interactions with any and all AI tools, IDEs, chatbots.... Now if it supports MCP you can with https://mem0.ai/ Give Claude desktop memory.

基本情報

カテゴリ

AI とエージェント

トランスポート

stdio

公開者

ryaker

投稿者

Richard Yaker

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "mem0-memory-general": {
      "command": "<PATH TO SERVER>/mcp-mem0-server",
      "args": [],
      "env": {
        "MEM0_API_KEY": "<YOUR MEM0 KEY>"
      }
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Mem0 General?

Mem0 General is a Model Context Protocol (MCP) server that bridges AI assistants with Mem0.ai’s persistent memory system. It enables compatible AI models to store, retrieve, search, and manage memories across short-term and long-term types, including knowledge graphs.

How to use Mem0 General?

Install the package with pip install mcp-mem0-server, then add a configuration entry to your MCP config file (e.g., ~/.cursor/mcp.json) that sets the MEM0_API_KEY environment variable and runs the command mcp-mem0-server. After restarting the client application (Cursor or Claude Desktop), the server tools are available.

Key features of Mem0 General

  • Modular architecture for maintainability and extensibility
  • Short-term memories: conversation, working, attention
  • Long-term memories: episodic, semantic, procedural
  • Semantic similarity search across stored memories
  • Custom memory categories and processing instructions
  • Selective memory filtering with include/exclude patterns

Use cases of Mem0 General

  • Give an AI assistant persistent recall of user preferences and past interactions
  • Build a knowledge graph from user-provided facts and relationships
  • Maintain conversational context across multiple sessions using short-term memory
  • Apply selective memory patterns to store only relevant information
  • Provide feedback to improve memory quality over time

FAQ from Mem0 General

How do I install and configure the Mem0 General server?

Run pip install mcp-mem0-server. Then add a configuration entry in your MCP config file (e.g., ~/.cursor/mcp.json) that uses "command": "mcp-mem0-server" and includes the environment variable MEM0_API_KEY set to your Mem0 API key.

What memory types are supported?

Short-term memories: conversation memory, working memory, and attention memory. Long-term memories: episodic memory, semantic memory, and procedural memory.

What is the default user ID?

All memories in the system use "default_user" as the default user_id unless overridden.

What advanced features are available?

Custom memory categories, memory processing instructions, knowledge graph relations between entities, selective memory filtering, and a feedback mechanism for memory quality.

Where does the memory data live?

Memories are stored and managed by Mem0.ai’s cloud memory system; the server acts as a client to that API using your Mem0 API key.

コメント

「AI とエージェント」の他のコンテンツ