
mcp-memory
by Puliczek·★ 143·Score 42
MCP Memory server that enables cross-conversation user memory using vector search technology on Cloudflare infrastructure.
Overview
MCP Memory is a specialized MCP server that provides persistent memory functionality for AI clients like Cursor, Claude, and Windsurf. It uses vector embeddings and similarity search to remember user preferences and behaviors across conversations. Built with Cloudflare Workers, D1 Database, and Vectorize, it implements a sophisticated architecture that stores both vector embeddings and original text in separate databases for efficient retrieval while maintaining data persistence. The server supports isolated namespaces per user, rate limiting, and TLS encryption to ensure security.
Try asking AI
After installing, here are 5 things you can ask your AI assistant:
When to choose this
Choose this MCP Memory server when you need persistent user memory across conversations for personalized AI interactions, especially if already using Cloudflare services.
When NOT to choose this
Avoid this if you need on-premises hosting, require different vector database providers, or want to avoid Cloudflare vendor lock-in.
Tools this server exposes
3 tools extracted from the READMEstore_memoryStore information about a user for later retrieval
retrieve_memoriesFind relevant memories based on semantic similarity
delete_memoryRemove a specific memory from storage
Note: Inferred from the architecture description and usage examples, though explicit tool names aren't directly documented in the README
Comparable tools
Installation
Installation Options
Option 1: One-Click Deploy
Click the deploy button and choose:
- Dimensions: 1024
- Metric: cosine
Option 2: Using the Template
- Click 'Use this template' button
- Clone the new repository
- Follow setup instructions
Option 3: Cloudflare CLI
npm create cloudflare@latest --git https://github.com/puliczek/mcp-memorySetup (Options 2 & 3)
- Install dependencies:
npm install- Create Vectorize index:
npx wrangler vectorize create mcp-memory-vectorize --dimensions 1024 --metric cosine- Run locally:
npm run dev- Deploy:
npm run deployClaude Desktop Configuration
Add to Claude Desktop config.json:
{
"mcpServers": {
"mcp-memory": {
"command": "npx",
"args": ["@puliczek/mcp-memory"]
}
}
}FAQ
- Can I use memory.mcpgenerator.com to store my memories?
- Yes, you can use memory.mcpgenerator.com to store and retrieve your memories for free. Your memories are securely stored and accessible only to you, but availability cannot be guaranteed.
- Can I run it locally?
- Yes, you can run MCP Memory locally for development using 'wrangler dev'. However, some features like vector search require a connection to Cloudflare's services.
Compare mcp-memory with
Last updated · Auto-generated from public README + GitHub signals.