CleanBook — Smart Bookmark Cleaning & Classification

English | 简体中文 | Docs

CleanBook is a bookmark cleaning and classification tool with an offline-first pipeline: rules and local ML by default, optional OpenAI-compatible LLM when you need semantic assistance.

Repository Overview

• CLI-first workflow with cleanbook and cleanbook-wizard
• Configurable taxonomy, rules, and thresholds in config.json
• Multi-format export to HTML, Markdown, and JSON
• Dedicated docs site for quick start, architecture, and development workflow

Quick Start

pipx install .
cleanbook -i examples/demo_bookmarks.html -o output

Run directly from source if you do not want an isolated CLI install:

python main.py -i examples/demo_bookmarks.html -o output

Read Next

• Quick Start
• System Architecture
• Development Guide

License

MIT — see LICENSE.