Skip to main content
The Model Context Protocol (MCP) server provides a standardized interface that allows any compatible AI model or agent to access your Raindrop.io data in a simple and secure way.

Server address

Our MCP server uses Streamable HTTP transport with OAuth 2.1 authentication at the following address:
On first connection your browser will open to authorize access to your Raindrop.io account.

Supported clients

Claude

claude.ai, Desktop, and Claude Code

ChatGPT

Plus, Pro, Business, Enterprise, Education

VS Code

GitHub Copilot

Cursor

AI code editor

Windsurf

AI code editor

Zed

Code editor (via mcp-remote bridge)

Other MCP clients

Hundreds of other tools now support MCP servers. You can configure them to use Raindrop.io MCP server with the following settings:
  • Command: npx
  • Arguments: -y mcp-remote https://api.raindrop.io/rest/v2/ai/mcp
  • Environment: None
For cloud-based or other hosted clients not listed on this page, please reach out to info@raindrop.io so we can enable support.

Available tools

The server exposes the following tools. Read-only tools only fetch data; the rest modify your library.

Bookmarks

Collections

Tags

Highlights

Account

FAQ

Run the following in Terminal to clear saved auth info, then try again:
If the issue persists, make sure you have the latest version of Node.js installed.
Yes, at the https://api.raindrop.io/rest/v2/ai/mcp endpoint. This is the only supported transport.
Yes. You can pass your Raindrop.io API token directly in the Authorization: Bearer <your-token> header instead of using the interactive OAuth flow.
Try this configuration instead: