I rarely call multiple AI agents at the same time, because I often get the context wrong (such as talking about the needs of project B in project A), and I feel that the parallel ability of the human brain is very limited.
You need to distinguish between the model and the agent. claude.md skills rules are handled by the agent.
The model is similar to the brain, and the agent is similar to the hands. The brain directs the hands to take claude.md skills rules and other resources to feed the brain.
If there is something that, once you start doing it, you often lose yourself in it—for example, before you know it, you’re busy until it gets dark—then that is the thing you truly passionate about.
I have developed an application to manage cards or documents on my mobile phone, which can quickly add authorization watermarks to cards or documents.
And a tool to combine some windows on the computer has also been developed, which can quickly open multiple specified application windows according to the set layout through shortcut keys.
There is also an application that meets my own bookkeeping habits.
Indeed, major platforms are becoming more and more closed, and many platforms have begun to not provide or delete website functions, allowing users to download apps before using them.
As for security-related platforms, I will only turn on those protection capabilities, such as banks, and other websites are not suitable!
Let's trace things back to their essence: when a user downloads a shopping app, their original intention is not to "use the software" but to "buy things." When we open a search engine, it's not to admire the web interface, but to "get answers."
For a long time, due to technological limitations, we had no choice but to endure complex intermediate ecosystems—software, websites, apps—to indirectly fulfill our ultimate goals.
But in the near future, this underlying logic will be fundamentally restructured. We could fulfill our needs directly through AI, bypassing all intermediate steps entirely.
