reMarkable

Provides access to the reMarkable library for AI tools, enabling browse, search, OCR, and note extraction across documents.

python

GitHub Stars

python

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "wavyrai-rm-mcp": {
      "command": "uvx",
      "args": [
        "rm-mcp@latest"
      ],
      "env": {
        "REMARKABLE_TOKEN": "<paste token from step 2>",
        "REMARKABLE_COMPACT": "1",
        "REMARKABLE_CACHE_TTL": "60",
        "REMARKABLE_PAGE_SIZE": "8000",
        "REMARKABLE_ROOT_PATH": "/Work",
        "REMARKABLE_INDEX_PATH": "~/.cache/rm-mcp/index.db",
        "REMARKABLE_OCR_BACKEND": "sampling",
        "REMARKABLE_INDEX_REBUILD": "1",
        "REMARKABLE_BACKGROUND_COLOR": "#FFFFFF",
        "REMARKABLE_MAX_OUTPUT_CHARS": "50000",
        "REMARKABLE_PARALLEL_WORKERS": "5"
      }
    }
  }
}

This MCP server enables you to access your reMarkable library from AI tools, letting you read, search, and navigate documents (including handwriting via OCR) as a seamless second brain for your workflows.

How to use

You can connect your reMarkable library to your AI tools to read and search across all documents, including handwritten notes converted with OCR. Use this server with Claude, VS Code Copilot, or other MCP clients to browse folders, read text, extract content from PDFs and EPUBs, and perform smart searches across your entire library. Configure your client with your access token, then start the MCP server locally so your AI tools can query your library through the MCP interface.

How to install

Prerequisites: you need the MCP client tooling available for your setup and access to the reMarkable Cloud via your Connect subscription.

One-command setup (recommended) uses the MCP client to prepare and configure access automatically.

uvx rm-mcp --setup

This opens your browser, prompts for the one-time code, and prints the ready-to-paste config for Claude Code and Claude Desktop.

Manual setup

If you prefer to configure manually, follow these steps to obtain a one-time code, convert it to a token, and wire the MCP client into Claude.

Get a One-Time Code Go to my.remarkable.com/device/browser/connect and generate a code.
Convert to Token Use the following command to register the code and obtain a token.

uvx rm-mcp --register YOUR_CODE

Claude Code configuration

Add the MCP server to Claude Code by supplying your token and selecting an OCR backend if needed.

claude mcp add remarkable \
  -e REMARKABLE_TOKEN='<paste token from step 2>' \
  -e REMARKABLE_OCR_BACKEND=sampling \
  -- uvx rm-mcp@latest

Claude Desktop configuration

To enable Claude Desktop access, add a local MCP configuration that points to the same token and the runtime command.

{
  "mcpServers": {
    "remarkable": {
      "command": "/Users/YOU/.local/bin/uvx",
      "args": ["rm-mcp@latest"],
      "env": {
        "REMARKABLE_TOKEN": "<paste token from step 2>"
      }
    }
  }
}

Run the MCP server locally

Start the MCP server using the runtime command shown in the manual setup to host the server locally for your MCP clients.

uvx rm-mcp@latest

Available tools

remarkable_read

Read and extract text from documents with pagination and search

remarkable_browse

Navigate folders in your library

remarkable_search

Search content across multiple documents

remarkable_recent

Get recently modified documents with previews

remarkable_status

Check connection status

remarkable_image

Get PNG/SVG images of pages with optional OCR output

Built by

VeilStrat

AI signals for GTM teams