Skip to main content

Backup and convert reMarkable tablet notebooks to PDF

Project description

RemarkableSync

RemarkableSync

GitHub GitHub release CI Build Executables PyPI version

A comprehensive Python toolkit for backing up reMarkable tablet notebooks, converting them to PDF, and transcribing handwriting to Markdown with AI — over USB or Wi-Fi.

[!IMPORTANT] Tested on reMarkable 2. Compatibility with reMarkable 1 is not guaranteed.

AI handwriting-to-text features require a GitHub Copilot or Claude (Anthropic) account. Support for additional providers can be added as needed.

Features

  • USB & Wi-Fi sync — connect via cable or wirelessly over your local network
  • Incremental backup — only downloads files that have changed (tracked by size, mtime, and MD5)
  • PDF conversion — v5 and v6 .rm formats with template backgrounds, folder hierarchy preserved
  • AI handwriting recognition — send page images to GitHub Models (GPT-4o) or Claude for transcription
  • Markdown export — each notebook becomes a .md file with YAML frontmatter and embedded page images
  • Watch mode — automatic periodic sync with system-tray status icon and run-at-startup option
  • Secure credential storage — SSH password and AI tokens stored in your system keyring

Quick Start

1. Install

# macOS (Homebrew)
brew tap jeffsteinbok/remarkablesync && brew install remarkablesync

# All platforms (pip)
pip install remarkablesync

# Or download a pre-built executable from the Releases page

2. Run the configuration wizard

RemarkableSync config

The wizard walks you through:

Setting Default
Connection mode USB or Wi-Fi (wizard can enable Wi-Fi SSH for you)
SSH password Saved to system keyring
Backup directory <AppData>/remarkablesync/backup (internal sync data)
PDF output ~/Documents/RemarkableSync/PDF
Markdown output ~/Documents/RemarkableSync/Markdown
AI provider GitHub Models (free) or Claude (requires API key)
AI token Stored securely in system keyring
Folders Choose which tablet folders to sync (or all)

[!TIP] Multi-device setup: Use the folder filter to sync different tablet folders to different computers — e.g., sync your "Work" folder to your work PC and "Home" to your personal machine. Each machine gets its own config.

[!TIP] Clear any directory field and press Enter to reset it to the default.

3. Run it

RemarkableSync watch

This will use your configured defaults and launch a sync cycle (backup → PDF → Markdown), then keep running and re-sync every 30 minutes. Check your output directories to verify everything looks right.

4. Set it to run at startup

Once you're happy with the output, enable run-at-startup from the system tray icon menu (or via the watch command). RemarkableSync will sync your tablet automatically in the background whenever your computer is on.

[!TIP] Obsidian users: Point the Markdown output directory at a folder inside your Obsidian vault. Your handwritten notes appear as searchable Markdown with embedded page images. Pair with Obsidian OneDrive Sync to sync your vault across devices.

AI Provider Setup

The AI handwriting-to-text features require either a GitHub Copilot account (for GitHub Models) or a Claude (Anthropic) API key. Support for additional providers can be added as needed.

Both AI provider SDKs are installed with pip install -r requirements.txt. The config wizard handles authentication interactively.

GitHub Models

The wizard runs a GitHub device-code flow to authenticate. No manual token setup needed.

Alternatively, set a GITHUB_TOKEN environment variable with a PAT that has models:read scope.

Claude (Anthropic)

  1. Go to https://console.anthropic.com/settings/keys
  2. Click Create Key and copy it (starts with sk-ant-api03-...)
  3. Paste it into the config wizard — it's saved in your system keyring

Default model: claude-sonnet-4-6.

Usage

After running config, most users only need watch. For one-off or scripted use:

# Default: backup + PDF conversion (uses saved config)
RemarkableSync

# Full pipeline: backup + PDF + AI OCR + Markdown
RemarkableSync md --with-backup --with-pdf

# Individual steps
RemarkableSync backup          # backup only
RemarkableSync convert         # PDF conversion only (from existing backup)
RemarkableSync md              # Markdown export only (from existing PDFs)

# Watch mode (periodic sync)
RemarkableSync watch           # uses saved config for interval, dirs, AI

Command Line Options

All commands read defaults from the saved config. CLI flags override config values.

Common:

  • -d, --backup-dir PATH — backup directory
  • -v, --verbose — debug logging
  • --version — version info

Connection:

  • --host HOST — tablet IP (default: 10.11.99.1)
  • --wifi — connect over Wi-Fi
  • --wifi-host HOST — tablet Wi-Fi IP (auto-discovered if omitted)

Backup:

  • -p, --password — SSH password (prompted if not saved)
  • --skip-templates — don't backup templates
  • --force-backup — re-download everything

Convert:

  • -o, --output-dir PATH — PDF output directory
  • --force-all — reconvert all notebooks
  • --sample N — convert first N notebooks only
  • --notebook NAME — convert a single notebook

Markdown (md):

  • -V, --vault-dir PATH — Markdown output directory
  • --ai-providergithub or claude
  • --ai-model — override default model
  • --ai-api-key — API key (prefer keyring or env vars)
  • --tags — comma-separated frontmatter tags (default: remarkable)
  • --no-images — skip embedding page images
  • --with-backup / --with-pdf — include earlier pipeline stages
  • --force-export — re-export all notes

Watch:

  • -i, --interval N — minutes between syncs (default: 30)
  • --systray / --no-systray — system tray icon (default: enabled)

Wi-Fi Connection

The config wizard can enable Wi-Fi SSH on your tablet automatically via USB. If you prefer to do it manually:

  1. Connect tablet via USB and SSH into 10.11.99.1
  2. Run rm-ssh-over-wlan on
  3. Find the IP: ip addr show wlan0
  4. Use that IP in the config wizard or --wifi-host

[!TIP] Assign a static DHCP lease to your tablet in your router so the IP doesn't change.

Generated Markdown Format

Each notebook becomes a folder with one Markdown file per page:

---
title: "My Meeting Notes"
source: reMarkable
remarkable_id: xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
notebook: Work
folder: Work/Meetings
page: 1
created: 2025-01-15
ai_provider: GitHubModelsProvider
ai_model: gpt-4o-mini
tags:
  - remarkable
---

## Action Items

- **Follow up** with Alice on the Q1 plan
- Schedule review for next Friday

![page 1](_images/page_001.png)

Page images are stored in a _images/ subfolder next to the Markdown files.

Troubleshooting

Problem Fix
Can't connect via USB Verify cable, ping 10.11.99.1, check SSH password
Can't connect via Wi-Fi Same network? Try ping remarkable.local or use --wifi-host <ip>
AI OCR returns errors Check API key/token, verify package installed (anthropic or openai), check rate limits
Empty Markdown files AI provider may have failed — check log file for details
Watch lock error Delete <backup-dir>/.remarkable_watch.lock
Permission errors Ensure output directories are writable; run as admin on Windows if needed

Security

  • SSH password and AI tokens are stored in your system keyring (never in plain-text config files)
  • All communication is over your local network (USB or LAN) — nothing goes to the internet except AI API calls
  • Config file at <AppData>/remarkablesync/config.json contains only non-secret settings

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

remarkablesync-2.0.0.tar.gz (93.5 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

remarkablesync-2.0.0-py3-none-any.whl (87.9 kB view details)

Uploaded Python 3

File details

Details for the file remarkablesync-2.0.0.tar.gz.

File metadata

  • Download URL: remarkablesync-2.0.0.tar.gz
  • Upload date:
  • Size: 93.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for remarkablesync-2.0.0.tar.gz
Algorithm Hash digest
SHA256 21d079eb842f7fd569e6ea883425c45ab664af420b1ccc30f2409554057facc1
MD5 328a17acae0ce6f46e5d5f49e3e72e6e
BLAKE2b-256 f5d1c761ee57b3322280d0edc1cbe9e12304a438a0c76dec8690f0957960a2d5

See more details on using hashes here.

Provenance

The following attestation bundles were made for remarkablesync-2.0.0.tar.gz:

Publisher: publish-pypi.yml on JeffSteinbok/RemarkableSync

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file remarkablesync-2.0.0-py3-none-any.whl.

File metadata

  • Download URL: remarkablesync-2.0.0-py3-none-any.whl
  • Upload date:
  • Size: 87.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for remarkablesync-2.0.0-py3-none-any.whl
Algorithm Hash digest
SHA256 8cd55e9a7e4f2eb0a299ab77821f8730899f846cdb7f08710c2544e91437a525
MD5 c14235362e275f05061a6656489f1dc8
BLAKE2b-256 531cdde03ce168353a9fc2d41f875093e30294e2860a7580d2abae2048851328

See more details on using hashes here.

Provenance

The following attestation bundles were made for remarkablesync-2.0.0-py3-none-any.whl:

Publisher: publish-pypi.yml on JeffSteinbok/RemarkableSync

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page