Recently, because the originally bound Xiaohongshu mobile phone number is no longer used, I plan to export the favorite notes in my old account in batches, but I have not been able to find a useful tool. Speaking of which, have you ever had similar troubles? I have more and more favorites, and every time I want to find inspiration notes, I have to flip through them for half a day, but I still can’t find them.
Collecting is like a “black hole”, which can only enter and exit, and over time it becomes an information dump full of debris……
I was also bothered by this problem for a long time until I used AI and finally solved it! So I made a small red book favorites one-click export widget, which can automatically export all the note data in your favorites, supporting note titles, authors, likes, note links, image links and other information, and the effect looks like this:
Media error: Format(s) not supported or source(s) not found
Download file: https://www.woshipm.com/wp-content/uploads/2025/05/lhFsfGR2BZKoOlsnitjn.mp4?_=300:0000:00Use the up/down arrow keys to increase or decrease the volume.
One-click export Xiaohongshu favorites effect video
Collecting is ≠ useful, what is the problem?
Xiaohongshu’s collection function is really easy to use, click to collect, create a folder, and complete it in one second. But:
the number of folders is too large, and the hierarchy is chaotic;
The content is mixed, life inspiration, work inspiration, artistic creation, decoration suggestions, etc.
The search function is not powerful enough, so I found an old collection point for half a day.
Over time, you will find: “I just collect it out of habit, but I have never looked back at them.” ”
So how do I consume the data of these favorites?
First thing I did:Export the content of Xiaohongshu favorites and use AI to do “moving + sorting” processing.
What does a product manager need to do?
In the process of a product from scratch, it is not easy to do a good job in the role of product manager, in addition to the well-known writing requirements, writing requirements, writing requirements, there are many things to do. The product manager is not what you think, but will only ask you for trouble, make a request:
View details >
To implement the information in the above multidimensional table, there are a few steps in a few simple words:
Scrape collection data
Through third-party plugins to extract the favorite links, titles, covers, and introductions, I used AI programming methods to implement a JavaScript script for me, and I just copied it directly and installed it directly in the Oil Monkey script. Of course, I am also very willing to share this useful tool of my own creation with you for free, if you also need to get the same tool, you can directly reply to “favorites” in the background of the WeChat public account to get the full source code to use.
Before that, you need to go to the Chrome app store to download a “tampering monkey”, download the address:
https://chromewebstore.google.com/detail/%E7%AF%A1%E6%94%B9%E7%8C%B4/dhdgffkkebhmkfjojejmpbldmpobfkfo?hl=zh-CN&utm_source=ext_sidebar
Tamper Monkey Chrome Download Address Page After Chrome browser is installed, then add the script file exported from this Xiaohongshu favorite to the tampered inside.
Add a new script to Tamper Monkey
Add the new script we provide
Then enter the favorites interface of your Xiaohongshu web version, click “Auto Scroll”, you can dynamically see the number of notes currently obtained, and finally export it with one click, isn’t it very convenient?
You can see the favorites plugin displayed
Import the knowledge base
Finally, import the organized content into the Feishu document, select the knowledge base with a “multi-dimensional table” structure, and truly become a “usable inspiration library”.
Upload data to a multi-dimensional table
Content classification/tag completion
Let AI automatically identify content topics, extract keywords, etc., to facilitate later search.
I add a column of “field shortcuts” to the multi-dimensional table, and then use “custom AI content” to fill in, the prompt word I mainly write is to let the AI recognize the title to generate keywords, there is still a lot of room to play, we will give you a detailed explanation in the next Feishu multi-dimensional table + field shortcut and other cases, it is a very good and very convenient tool.
Add AI autofill to field shortcuts
How good is the AI tidying up?
For example, I used to have 1000+ notes in my Xiaohongshu favorites, but now I use AI to process them like this:
AI categorized categories are automatically generated
Each category is clear, and even the title, abstract, keywords, and original links are retained, making it particularly convenient to search and filter. At the same time, if this data can be learned and referenced by large models as knowledge base data, you will be able to search for related questions and generate relevant content more vertically and accurately! Of course, there are also very big uses in this, which we will slowly use in the subsequent collection of various practical cases of Xiaohongshu, such as reading the link of the article and analyzing the content of the text, imitation, disassembly, etc., and we will share it step by step later.
Don’t let inspiration sink and turn them into your assets
The collection itself is correct, the point isMake it available, checkable, and inspiring。
What AI brings is not a show-off skill, but a good tool that can really solve “information anxiety”.
So, stop letting your favorites become a “forgotten corner”, try to use AI to “move and organize”, maybe you will rediscover a lot of the shining points of your collection at that time.
