It is divided into the following main sections: background information, task requirements and output formats
1. Background information, all information that is useful for its generation but not accessible, such as
– 付费文章（它访问不到的）
– 视频文字稿（它不能看视频）
– 图片或 PDF（作为附件即可）
– 其他任何对于生成有帮助的信息
Note that background information should be wrapped in XML if there is a lot of it, otherwise the AI won't be able to tell which is background information and which are commands, for example:
<background>
<news>
"Paid News"
</news>
"Paid Articles"
<transcript>
"Video Transcript"
</transcript>
</background>
2. What kind of report would you like it to make, as required by the mandate?
– 分析的主题是什么？
– 希望它检索的信息源（比如英文、权威信息源、学术论文、政府网站、GitHub等等）
– 要分析主题下面的哪些点？
– 是深度解析还是摘要性质的
3. Output formats, what are the formatting requirements for reports
– 中文报告还是英文报告还是双语报告？
– 数据是否用表格展示？（它现在画不了图表，要求数据整理成表格即可）
English 
简体中文 
日本語 
Deutsch 
Português do Brasil