Skip to content
There’s something a bit hard to say.
Actually, I’ve been using Baidu’s homepage navigation feature for over a decade, which is this thing:
Because hao123 is too flashy, and as a long-time Tieba user, using Baidu’s navigation is indeed more convenient. After years of using it, I have accumulated over a hundred favorite websites, and the migration cost is too high, so even though later on I could deploy many great navigation applications via Docker on NAS, and despite Baidu’s forced inclusion of a bunch of junk news and Baijiahao that I don’t want to see in the navigation panel, as long as it doesn’t affect usage, I just put up with it.
But recently, for some unknown reason, Baidu revamped its interface, and the entire navigation module became narrower, showing only 6 icons per row, leaving a huge blank area displaying meaningless background images. Now my 100+ websites are stacked into several screens.
At first, I thought there was a problem with the settings, but after searching for a long time, I couldn’t find anything. I searched online and found that many people were complaining.
It’s really unbelievable.
This time, I decided not to sit idly by.
Who knows what other weird things Baidu’s ** designers will come up with.
Abandon Baidu and switch to a new navigation website.
Actually, there were alternatives before.
But after all these years, I’ve accumulated too many navigation websites, and just thinking about re-entering them one by one is exhausting.
But times have changed; in the age of AI, we can use AI methods to solve problems that were previously difficult to tackle.
First, we can use AI to extract the original navigation categories and websites.
The free DeepSeek-R1 can accomplish this:
Right-click on the navigation section in Baidu—Inspect,
Then copy the entire front-end code of the section.
Just tell the AI, this is the front-end code of a navigation page, please help me extract the categories, website names, and URLs from it,
It can automatically extract the categories, names, and links of the websites.
Next, you can let AI complete this navigation page.
You can do this in a chat tool:
It’s recommended to use Claude, which allows for repeated modifications and adjustments in Artifacts and supports previews.
If you need to make changes, just continue the conversation.
The back-end implementation is also done by generating code through chat.
Special reminder: If, like me, you have no coding background, it is recommended to ask it to implement it using Python or PHP, which are relatively easy to deploy.
However, chat tools always require copying and pasting, and you have to find the code locations everywhere, which is quite troublesome. For a more automated solution, you can also use AI programming tools:
If you seek extreme cost-effectiveness, you can use VS Code (actually, Cursor will suffice) with Roo Code or Cline plugins, connecting to DeepSeek’s API; writing a simple page is sufficient.
The usage method is the same as directly using the chat tool, it’s just a conversation.
I completed it directly through Cursor, and the Composer function of Cursor can also automatically write code across files and debug, using the Claude 3.5 Sonnet model that I am very familiar with.
Even during the development process, I had it write a small tool to ping and send HTTP requests to all 150+ websites I accumulated over the past decade, excluding a dozen that were already down.
The final result is shown in the image:
-
I put it on my own Alibaba Cloud server (to be precise, I directly SSH into the cloud server to generate the code) and bound my own domain name.
-
It was designed to a width that I am quite satisfied with, and it can also adapt to small-screen devices.
-
While cleaning up invalid websites, I also re-categorized them.
-
I added a simple back-end, and after migrating the original Baidu navigation websites, it supports editing and adding new websites, as well as drag-and-drop layout. (My back-end does not use a database; I simply used a Python file to store these websites, so I can also edit this file from the server backend to make modifications and layouts.)
-
I added time and date, weather functions, and quick search boxes for three major search engines.
-
I added entry points for NAS and AI secondary aggregation pages, which can open the Docker applications I deployed on NAS and my commonly used AI tools.
-
I added an auto-saving notepad. This way, when copying and pasting text between multiple Windows, Mac, and mobile devices, I no longer have to be angry with the sometimes unreliable WeChat input method, nor do I need to log into multiple WeChat or QQ accounts. I just paste into this box, and when I open the homepage of the browser on another device, I can quickly copy it.
In short, by being diligent, I said goodbye to Baidu navigation and feel much more comfortable.
