🤝 응원을 표현하세요 - 내용이 마음에 드셨다면 알려⭐️주세요

---

MCP 메모리

MCP 메모리는 MCP 서버로, **MCP 클라이언트(커서, 클로드, 윈드서프 등)**에게 **대화 전반에 걸쳐 **사용자 정보(선호도, 행동)를 기억할 수 있는 **능력을 제공합니다. 벡터 검색 기술을 사용해 단순한 키워드가 아니라 의미를 바탕으로 관련 기억을 찾습니다. Cloudflare Workers, D1, Vectorize(RAG), Durable Objects, Workers AI, Agents로 구축되었습니다.

📺 비디오

🚀 한번 해봐

https://memory.mcpgenerator.com/

🛠️ 자신만의 MCP 메모리 배포 방법

옵션 1: 원클릭으로 자신만의 MCP 메모리를 Cloudflare에 배포

벡터화 만들기 섹션에서 다음을 선택하세요:

• 치수: 1024
• 미터법: 코사인

클릭 버튼 “생성 및 배포”

Cloudflare 대시보드에서 "Workers and Pages"로 가서 방문을 클릭하세요

옵션 2: 이 템플릿을 사용하세요

1. 이 저장소 상단의 “이 템플릿 사용” 버튼을 클릭하세요
2. 새 저장소 복제
3. 아래 설정 지침을 따르세요

옵션 3: CloudFlare CLI로 만들기

🔧 설정 (옵션 2와 3만 있음)

1. 의존성 설치:

2. 벡터화 인덱스 생성:

3. 랭글러 설치:

---

MCP Memory

MCP Memory is a MCP Server that gives MCP Clients (Cursor, Claude, Windsurf and more) the ability to remember information about users (preferences, behaviors) across conversations. It uses vector search technology to find relevant memories based on meaning, not just keywords. It’s built with Cloudflare Workers, D1, Vectorize (RAG), Durable Objects, Workers AI and Agents.

📺 Video

🚀 Try It Out

https://memory.mcpgenerator.com/

🛠️ How to Deploy Your Own MCP Memory

Option 1: One-Click Deploy Your Own MCP Memory to Cloudflare

In Create Vectorize section choose:

• Dimensions: 1024
• Metric: cosine

Click button “Create and Deploy”

In Cloudflare dashboard, go to “Workers & Pages” and click on Visit

Option 2: Use this template

1. Click the “Use this template” button at the top of this repository
2. Clone your new repository
3. Follow the setup instructions below

Option 3: Create with CloudFlare CLI

npm create cloudflare@latest --git https://github.com/puliczek/mcp-memory

🔧 Setup (Only Option 2 & 3)

1. Install dependencies:

npm install

2. Create a Vectorize index:

npx wrangler vectorize create mcp-memory-vectorize --dimensions 1024 --metric cosine

3. Install Wrangler:

npm run dev

4. Deploy the worker:

npm run deploy

🧠 How It Works

1. Storing Memories:

   • Your text is processed by Cloudflare Workers AI using the open-source @cf/baai/bge-m3 model to generate embeddings
   • The text and its vector embedding are stored in two places:
     • Cloudflare Vectorize: Stores the vector embeddings for similarity search
     • Cloudflare D1: Stores the original text and metadata for persistence
   • A Durable Object (MyMCP) manages the state and ensures consistency
   • The Agents framework handles the MCP protocol communication

2. Retrieving Memories:

   • Your query is converted to a vector using Workers AI with the same @cf/baai/bge-m3 model
   • Vectorize performs similarity search to find relevant memories
   • Results are ranked by similarity score
   • The D1 database provides the original text for matched vectors
   • The Durable Object coordinates the retrieval process

This architecture enables:

• Fast vector similarity search through Vectorize