Skip to content
BSWEN
Blog

How to Use Supermemory MCP Server with Claude and Cursor

Cowrie
Cowrie
Dev @ Bswen

Problem

I use Claude and Cursor for coding, but every new conversation starts from scratch. I tell Claude about my project structure, my preferences, and my tech stack. Then I open a new chat and it is all gone. I repeat myself constantly.

I wanted a way to give my AI assistants long-term memory without switching tools or building a custom integration.

What Is MCP?

MCP (Model Context Protocol) is a standard that lets AI tools call external services. Supermemory provides an MCP server that adds persistent memory to any compatible assistant. Supported clients include Claude, Cursor, Windsurf, VS Code, Claude Code, OpenCode, OpenClaw, and Hermes.

One-Line Install

The fastest way to get started:

Terminal
npx -y install-mcp@latest https://mcp.supermemory.ai/mcp --client claude --oauth=yes

Replace claude with cursor, windsurf, vscode, or another supported client. The command registers the MCP server and handles OAuth automatically.

Manual Configuration

If you prefer to configure manually, add this to your client’s MCP settings:

mcp-config.json
{
  "mcpServers": {
    "supermemory": {
      "url": "https://mcp.supermemory.ai/mcp"
    }
  }
}

With API key authentication instead of OAuth:

mcp-config-api.json
{
  "mcpServers": {
    "supermemory": {
      "url": "https://mcp.supermemory.ai/mcp",
      "headers": {
        "Authorization": "Bearer sm_your_api_key_here"
      }
    }
  }
}

Three Built-In Tools

Once connected, your assistant gets three tools:

  • memory — save or forget information
  • recall — search memories by query, returns relevant memories plus user profile
  • context — inject the full profile into the current conversation

In Cursor or Claude Code, type /context to trigger the context tool. This pulls your entire profile into the chat so the assistant knows who you are and what you are working on.

How It Works in Practice

Here is a real workflow I use:

  1. I tell Cursor: “I am building a Next.js app with Stripe integration.”
  2. Supermemory extracts and stores that fact automatically.
  3. The next day, I open a new chat and say: “Continue working on that project.”
  4. Cursor recalls: “You are building a Next.js app with Stripe integration.”

No setup required beyond the initial install.

Scoping by Project

I use container tags to separate work and personal context. When I start a work conversation, I mention the project name. Supermemory scopes memories accordingly, so my personal side projects do not leak into work chats.

MCP vs SDK

One thing confused me at first: the MCP server is for end users who want memory in existing AI tools. The SDK is for developers building custom agents. If you are using Claude or Cursor, use MCP. If you are building your own app, use the SDK or framework integrations.

Summary

In this post, I showed how to add persistent memory to Claude and Cursor using the Supermemory MCP server. The key point is that one install command gives your AI assistant three memory tools, so it remembers your preferences, projects, and past discussions across every conversation.

Final Words + More Resources

My intention with this article was to help others share my knowledge and experience. If you want to contact me, you can contact by email: Email me

Here are also the most important links from this article along with some further resources that will help you in this scope:

Oh, and if you found these resources useful, don’t forget to support me by starring the repo on GitHub!

Comments

Tags: