Read an EPUB book chapter by chapter in a local web browser.
Copy chapter text easily to discuss or summarize with an AI assistant.
Keep a small personal library of EPUB books served from your own computer.
| karpathy/reader3 | pgadmin-org/pgadmin4 | markfzp/act-plus-plus | |
|---|---|---|---|
| Stars | 3,612 | 3,611 | 3,615 |
| Language | Python | Python | Python |
| Setup difficulty | easy | moderate | hard |
| Complexity | 1/5 | 3/5 | 5/5 |
| Audience | vibe coder | developer | researcher |
Figures from each repo's GitHub metadata at analysis time.
Built quickly for personal use and not actively maintained by the author.
Reader3 is a small, self-hosted web app for reading EPUB books one chapter at a time on your own computer. The main idea behind it is convenience when reading alongside an AI assistant: because it shows one chapter at a time, you can easily select the text, copy it, and paste it into a conversation with a language model to discuss, summarize, or ask questions about what you just read. To use it, you point it at an EPUB file from the command line. It processes the book and registers it to a local library folder. Then you start a local web server and visit it in your browser to see your library and read. Adding more books means running the same command with different files, removing a book means deleting its folder. The project uses uv, a Python tool that handles dependencies without requiring a separate installation step. The README gives a short example using a free copy of Dracula downloaded from Project Gutenberg. The author, Andrej Karpathy, describes this as something he built quickly for personal use, roughly 90 percent written with AI assistance, and shared as an example of how easy it is to build small tools this way. He is explicit that he does not intend to maintain or improve it. The code is provided as-is for others to read, run, or modify however they like.
A small self-hosted app that lets you read EPUB books one chapter at a time, made for easily copying text into an AI chat to discuss what you just read.
Mainly Python. The stack also includes Python, uv.
No license information is stated in the README.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly vibe coder.
This repo across BitVibe Labs
Verify against the repo before relying on details.