AI 分析时，
插件会保留每个卡片的属性参数，并以 JSON 的形式将一篇（或多篇）文献的注释喂给 AI
，比如以下 JSON 表示的是来自两篇文献的注释（为方便举例，我随便标注了几条）：
{ "ref-1": { "title": "Cheeke et al. 1999 - Acoustic wave gas sensors.pdf", "annotations": [ { "annotationType": "highlight", "annotationColor": "#ffd400: 黄色", "annotationText": "One reason is that such sensors are small, inexpensive, sensitive and they respond to virtually all gases.", "annotationComment": "声表面波气体传感器", "annotationTags": "#SAW gas sensor/优势" }, { "annotationType": "highlight", "annotationColor": "#e56eee: 洋红色", "annotationText": "Types of acoustic wave sensors", "annotationComment": "声表面波传感器的主要类型", "annotationTags": "#SAW Sensor/主要类型" } ] }, "ref-2": { "title": "Jakubik et al. 2011 - Surface acoustic wave-based gas sensors.pdf", "annotations": [ { "annotationType": "highlight", "annotationColor": "#ffd400: 黄色", "annotationText": "A surface acoustic wave (SAW) is an acoustic wave traveling along the surface of a material exhibiting elasticity, with an amplitude that typically decays exponentially with depth into the substrate.", "annotationComment": "声表面波的定义", "annotationTags": "#SAW/定义" }, { "annotationType": "highlight", "annotationColor": "#ffd400: 黄色", "annotationText": "Rayleigh waves have a longitudinal and vertical shear component that can couple with any media in contact with the surface —Fig. 1.", "annotationComment": "瑞利波", "annotationTags": "#SAW/Rayleigh wave" } ] } }
从中我们可以看出：
🎈
每篇文献都有一个编号，格式为
ref-xxx
，以及相应的标题参数
title
。如此一来，AI 分析笔记时可以轻松溯源。
参数
annotationType
、
annotationColor
、
annotationText
、
annotationComment
、
annotationTags
代表了注释的类型、颜色、文本、评论、标签。这些参数给 AI 分析提供了很大空间，比如你可以让 AI 单独分析某个参数。
