MCP Memory Server
A simple memory storage server for Claude using the Model Context Protocol (MCP), enabling Claude to store and retrieve text memories across conversations.
README
MCP Memory Server
A simple memory storage server for Claude using the Model Context Protocol (MCP). Perfect for hobby projects or learning how to build your first MCP server!
This server lets Claude remember things across conversations by storing and retrieving memories with optional tags.
What It Does
Store text memories that Claude can search and retrieve later. Great for:
- Remembering user preferences
- Storing project notes
- Building knowledge bases
- Learning MCP development
Available Tools
store_memory: Store a memory with optional tagsretrieve_memories: Search memories by contentlist_memories: Show all stored memoriesdelete_memory: Delete a specific memoryclear_memories: Delete all memories
Quick Start
1. Install & Run
# Clone and setup
git clone https://github.com/imyashkale/mcp-memory-server.git
cd mcp-memory-server
npm install
npm start
Server runs on http://localhost:3000
2. Add to Claude
claude mcp add memory http://localhost:3000/message --transport http
That's it! Claude can now use the memory tools.
Basic Usage
Once connected to Claude, you can:
# Store memories
"Remember that I like both dogs and cats"
# Retrieve memories
"What do you remember about my pet preferences?"
# List all memories
"Show me all my stored memories"
Development
Run in dev mode (auto-restart)
npm run dev
Project Structure
index.js- Main server file with MCP toolspackage.json- Dependencies and scripts- In-memory storage (resets on restart)
Environment Variables
PORT- Server port (default: 3000)LOG_LEVEL- Logging: error, warn, info, debug (default: info)
Memory Format
Each memory contains:
id- Unique numbercontent- The memory texttags- Optional categorization tagstimestamp- When it was created
Docker (Optional)
# Build and run
docker build -t mcp-memory-server .
docker run -d -p 3000:3000 mcp-memory-server
# Add to Claude with Docker
claude mcp add memory --transport http -- http://localhost:3000/message
What's Next?
This is a basic MCP server to get you started. You can extend it by:
- Adding persistent storage (database)
- Adding more search features
- Building a web interface
- Adding authentication
Troubleshooting
Port in use? Change the port:
PORT=3001 npm start
claude mcp add memory http://localhost:3001/message --transport http
Connection issues? Check the server is running:
curl -X POST http://localhost:3000/message -H "Content-Type: application/json" -d '{"jsonrpc": "2.0", "id": 1, "method": "tools/list"}'
Contributing
This is a hobby project, but contributions are welcome! Feel free to:
- Fork the repository
- Create a feature branch
- Make improvements
- Submit a pull request
License
MIT License - feel free to use this project for learning, hobby projects, or commercial use.
Support
- Issues: Open an issue on GitHub for bugs or feature requests
- Questions: Check the MCP documentation or start a discussion
- MCP Resources: Model Context Protocol Docs
Built with Node.js and the official MCP SDK. Perfect for learning MCP development!
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。
mcp-server-qdrant
这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。