四清单索引框架:自动汇总学习材料(如 Chrome 书签),用 reading 指标管理 backlog,并复用 tool-list / solved-list 加速消化。
仓库:github.com/fooSynaptic/quanta-learn
- 四清单:
tool-list→solved-list→reading-list→problem-list - 浏览器取材:只读导入书签 / 历史 / 会话 →
reading-list索引 - 本地看板:
python3 dashboard/server.py→ http://127.0.0.1:8765/
git clone https://github.com/fooSynaptic/quanta-learn.git
cd quanta-learn
pip install -r requirements.txt
bash scripts/init_local_catalog.sh| 文档 | 内容 |
|---|---|
| DESIGN.md | 消化闭环、数据流、路线图 |
| AGENTS.md | Agent 协议 |
| docs/UI-DESIGN.md | Dashboard 设计 |
| docs/TODO.md | 开发待办 |
| catalog/README.md | 本地索引初始化 |
export CHROME_USER_DATA_DIR="<your-browser-profile-dir>" python3 scripts/import_chrome_sources.py python3 scripts/classify_reading_items.py python3 scripts/reading_to_problem.py python3 scripts/sync_catalog_from_legacy.py python3 scripts/build_dashboard_stats.py
Python 3.10+,pip install -r requirements.txt