What It Does
I read a lot of papers on my reMarkable tablet. This toolkit automates the annoying parts: downloading papers from arXiv, classifying them by topic using Claude, and syncing them to the right folders on the device.
The core is pyrmapi, a Python wrapper around the rmapi Go binary. It handles binary management, authentication, and file operations. On top of that, there's a classification pipeline that reads PDFs, sends them to Claude for categorization, and organizes them automatically.
How It Works
- Pull papers from arXiv or other sources
- Claude classifies each paper by research area
- Papers get sorted into folders on the reMarkable
- TaskManager integration tracks what I've read and what's queued
Current State
Right now it's a Python wrapper around the rmapi binary. I'd like to eventually rewrite the cloud API bits in pure Python to drop the Go dependency, but the wrapper works fine for my needs.