supermemory

Supermemory MCP acts as a universal memory bridge that lets memories you create in one LLM be accessible across all other LLMs supported by your environment. By hosting a local MCP server with your API key, you enable any connected client to read and write memory entries without requiring separate logins or paywalls. The server relies on the Supermemory API to index and retrieve memories quickly, providing a seamless experience when switching between ChatGPT, Claude, or other LLMs. To use it, start the MCP server with your API key set in the environment, then point each LLM client to the local MCP endpoint so memories are centralized and portable. The one-command setup mentioned in the project notes is designed to minimize setup friction and get you running quickly.

Prerequisites:

• Node.js (recommended LTS version) and npm installed
• Access to a Supermemory API key

Steps:

1. Clone or download the MCP repository for Supermemory: git clone https://github.com/supermemoryai/supermemory cd supermemory

2. Install dependencies: npm install

3. Configure environment:

   • Create a .env file at the project root (or export the variable in your shell) and add: SUPERMEMORY_API_KEY=your_api_key_here

4. Run the MCP server: npm start

5. Connect clients:

   • Point your LLM clients to the MCP server URL (e.g., http://localhost:PORT or the deployed address) to access universal memories.

Notes and tips:

• The MCP is designed to be compatible across multiple LLM clients without requiring per-client logins.
• Ensure your SUPERMEMORY_API_KEY is kept secret and not exposed in client-side code or public repos.
• If you encounter connectivity issues, verify that the server is running and listening on the expected port, and that CORS/basic auth settings (if any) align with your client tooling.
• The project mentions a one-click DXT installer; if you plan to use that, ensure you trust the source and follow any accompanying documentation for environment setup.
• This MCP version is part of a deprecated v1 line in favor of newer implementations; consider migrating to the latest MCP version in the official repository for ongoing support.