更多的记录方式

[codexu]

NoteGen 天生支持了三种记录方式：截图、插图、文本。
未来可能增加的几种记录方式：

• 文件记录：上传文件，识别文件内容，例如记录 PDF、Word 等文本文件或 Excel，csv 等提供数据的文件。
• 链接记录：复制链接，爬取网页内容。
• 拍照记录：功能类似于插图记录，调用相机，更适合开会或上课场景，在移动端开发时作为主要记录手段。

[BradKML]

Seconding this, we need something that records, and something that converts MarkDown back into PDF or Word file ala Pandoc or some other similar tool.
When it comes to recording, there are many things to handle

1. parsing tables AKA Excel + CSV/TSV (weird use case but understandable)
2. Juggling PDF/TeX/EPUB/MOBI for both documents and scientific papers, and then chunking them logically (including treating index and appendix specially)
3. Crawling and/or archiving the web and extracting articles (ideally, whole-blog scraping for comprehensive info)
4. handling code files (kind of a niche use case,e assuming NoteGen will be MCP-compatible)
5. Not sure if audio/video transcripts will be handled since they require ASR tooling to be somewhat functional, can double as voice-to-note for free associative noting with chatbots

Related note #269