This server provides a powerful and flexible integration for leveraging reMarkable notes in AI workflows. It supports multiple connection modes, has comprehensive read/write tools, and offers smart features like auto-OCR and tag filtering. Active development and clear documentation add to its usability.
Strongest · Documentation 16/20 Weakest · Reliability 10/20
Static review · not runListed 2026-08-07
Read the FMRS scoring method →Runtime：Python
REMARKABLE_TOKEN
reMarkable Cloud API token; generate from a one-time code at my.remarkable.com/device/desktop/connect via `uvx remarkable-mcp --register`. Required for cloud mode.
GOOGLE_VISION_API_KEY
Google Cloud Vision API key for handwriting OCR; create it in the Google Cloud Console credentials page.
REMARKABLE_OCR_BACKEND
Select the OCR backend: sampling, google, tesseract, or auto (default).
REMARKABLE_READ_ONLY
Set to 1 to run the server in read-only mode, disabling all write tools.
REMARKABLE_SKIP_CONFIRM
Set to 1 to allow deletes without a prompt when the client lacks MCP elicitation support.
REMARKABLE_ROOT_PATH
Scope all server operations to a specific folder on the reMarkable, e.g. /Work.
REMARKABLE_BACKGROUND_COLOR
Default background color for page image rendering, #RRGGBB or #RRGGBBAA format, default #FBFBFB.
REMARKABLE_RETRY_ATTEMPTS
Maximum number of cloud API request attempts, default 3.
REMARKABLE_RETRY_DELAY
Base delay in seconds for exponential backoff, default 2.0.
REMARKABLE_SYNC_WORKERS
Number of parallel workers for cloud library fetches, default 16 (clamped to 64).
REMARKABLE_DISABLE_CACHE
Set to 1 to fully disable the on-disk blob cache in cloud mode.
REMARKABLE_CACHE_DIR
Blob cache directory for cloud mode, default ~/.remarkable/cache/blobs.
REMARKABLE_CACHE_MAX_BLOB
Blobs larger than this are streamed but not cached, default 4 MiB.
REMARKABLE_DISABLE_CLOUD_FALLBACK
Set to 1 to disable automatic fallback to cloud mode when the device is unreachable.
uvx remarkable-mcp --register YOUR_CODE.GOOGLE_VISION_API_KEY..vscode/mcp.json for VS Code or Claude Desktop's config.--usb, --ssh, or none for cloud).{
"mcpServers": {
"remarkable": {
"command": "uvx",
"args": ["remarkable-mcp", "--usb"],
"env": {
"GOOGLE_VISION_API_KEY": "your-api-key"
}
}
}
}
Works in clients that read an "mcpServers" key. For clients such as VS Code that use a different key, the configurator below converts it.
{
"servers": {
"remarkable": {
"command": "uvx",
"args": [
"remarkable-mcp",
"--usb"
],
"env": {
"GOOGLE_VISION_API_KEY": "your-api-key"
}
}
}
}
Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).
claude mcp add remarkable -e GOOGLE_VISION_API_KEY=your-api-key -- uvx remarkable-mcp --usb
Run it in a terminal; replace any <…> placeholders with your own values first.
The client's tool list should show tools like remarkable_read, remarkable_browse, and remarkable_status; calling remarkable_status confirms the active transport (USB/SSH/Cloud) and its capability matrix.
Once connected, you can ask your AI assistant things like:
No matching tools
Listed from the project's documentation, not tested by this site.
The reMarkable MCP Server gives AI assistants access to all notes and documents on your reMarkable tablet. It supports browsing folders, searching documents, and reading any file, including extracting text from handwritten notes via OCR. Multiple connection modes are offered: USB Web interface (recommended, no subscription or developer mode), SSH (requires developer mode), and Cloud (requires Connect subscription, works device-free). Write tools are included (upload, create folder, move, rename, delete), and document text and page images are available as MCP resources.
Comprehensive access to Google Search Console via MCP, with advanced analytics and quick win detection.
A production-ready MCP server that connects AI assistants to your Nextcloud instance.
An MCP server for interacting with Microsoft 365 and Office services through the Graph API.
Secure MCP access to Google Drive, Docs, Sheets, Slides, and Calendar
Source revision 2ad0e1751226 Data synced 2026-10-12 Read the FMRS scoring method