Mmp Mcp
What is Mmp Mcp
MMP-MCP (Model-Memory-Protocol MCP) is an open protocol implementation for managing AI model memory. It supports the storage and retrieval of structured memories in a tree format.
Use cases
Use cases include managing user preferences in chatbots, storing conversation history for personalized responses, and maintaining context in multi-turn dialogues.
How to use
To use MMP-MCP, you can install it via npm, yarn, or pnpm. Start the MCP server using the command ‘npx mmp-mcp’ or by importing and starting it in your code with ‘startMMPServer’.
Key features
Key features include the ability to create and manage memory collections, support for tree-structured memory storage, handling various content types (JSON, Markdown, XML, plain text), initialization management of memory nodes, and support for batch operations.
Where to use
MMP-MCP can be used in AI applications that require memory management across sessions, such as chatbots, virtual assistants, and any interactive AI systems that benefit from remembering user interactions.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Overview
What is Mmp Mcp
MMP-MCP (Model-Memory-Protocol MCP) is an open protocol implementation for managing AI model memory. It supports the storage and retrieval of structured memories in a tree format.
Use cases
Use cases include managing user preferences in chatbots, storing conversation history for personalized responses, and maintaining context in multi-turn dialogues.
How to use
To use MMP-MCP, you can install it via npm, yarn, or pnpm. Start the MCP server using the command ‘npx mmp-mcp’ or by importing and starting it in your code with ‘startMMPServer’.
Key features
Key features include the ability to create and manage memory collections, support for tree-structured memory storage, handling various content types (JSON, Markdown, XML, plain text), initialization management of memory nodes, and support for batch operations.
Where to use
MMP-MCP can be used in AI applications that require memory management across sessions, such as chatbots, virtual assistants, and any interactive AI systems that benefit from remembering user interactions.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Content
MMP-MCP (Model-Memory-Protocol MCP)
一种用于 AI 模型记忆管理的开放协议实现，支持树形结构化记忆的存储和检索。
介绍
MMP-MCP 是 Model-Memory-Protocol 的 MCP 实现，提供了一套完整的 API 接口，用于管理 AI 模型的记忆树。MMP 支持树形结构化记忆，可以方便地存储和检索 AI 模型在交互过程中产生的记忆，实现跨会话的记忆管理。
功能特点
- 创建和管理记忆集合
- 树形结构化记忆存储
- 支持多种内容类型（JSON、Markdown、XML、纯文本）
- 记忆节点初始化管理
- 批量操作支持
使用方法
{
"mcpServers": {
"memory": {
"command": "npx",
"args": [
"mmp-mcp"
],
"env": {
"MMP_RPC_ENDPOINT": "http://localhost:18080/rpc",
"MMP_DEFAULT_MEMORY_ID": "mm-xxxx"
}
}
}
}API 接口
MMP-MCP 提供了以下 API 接口：
memory.GetInitNodes: 获取所有待初始化节点memory.Add: 添加记忆memory.Get: 获取记忆memory.List: 获取记忆列表memory.Update: 更新记忆memory.Delete: 删除记忆memory.Batch: 批量获取记忆节点memManager.Create: 创建新的记忆树 IDmemManager.ApplyTemplate: 为记忆 ID 应用模板
详细的 API 文档请参考 OpenRPC 规范。
许可证
MIT
Dev Tools Supporting MCP
The following are the main code editors that support the Model Context Protocol. Click the link to visit the official website for more information.
