小唯 A06
|
dbf065bcfa
|
feat: page_dom_extract.py — text-based DOM simplifier
- HTML → LLM-readable simplified text format (page-agent inspired)
- [N]<tag>text</tag> numeric-indent notation
- Interactive element detection (a/button/input/select/textarea/details/summary/label + ARIA roles + contenteditable)
- curl primary fetch, lxml parsing, html.unescape for Chinese
- CLI: python3 page_dom_extract.py <url>
Ref: alibaba/page-agent d1
|
2026-07-09 16:58:11 +08:00 |