KaseKun

loved it!

> still have access to some decent model

Not really a requirement, if you did fine without the tools before you would do fine again. The difference is you would know which ones are worth manually maintaining

I have no use for this, however I am stunned by the simple and effective solution of using json+sse instead of an API to allow agent interaction Amazing work

You have to leave your local maximum to find the global maximum.

Code doesn't exist in a vacuum though.

Code bases that you "work in" (maintain, etc) solve real world problems, and solving those problems should trump cleanliness every time

Codebases that are clean are typically showcase examples that sit on a shelf to be admired and appreciated.

I have a (relatively well informed) view that people who aledge that "if you used an llm you did no thinking or problem solving" have never in fact used an a llm to generate anything particular complex.

You do indeed need to do quite a bit of thinking and problem solving, to build things with an llm.

If you disagree, repeat this project, so you can share with us how little thinking it required.

I can picture this comment at the 50th percentile on the midwit meme

On either side it says "I just ask the model why it did that"

I was stubbornly of the same mindset, but had friends and colleagues that raved about skills, i thought it was hype cycle context management - i'm happy to be proven wrong

haha interesting - the article is 99% written by me, but i had gemini review it and sharpen up the send off because it felt weak.

I guess this goes to show that even a subtle touch of an LLM can undermine authenticity.

edit: i've removed that line. I don't like to edit articles after publish (call me old fashioned, but i try to be honest and transparent), in this case though the line adds nothing and your call-out has taught me a good lesson: shit human writing is better than "good" AI writing.

Author here,

> It looks like someone just pointed a LLM at the codebase and asked it to write an article.

Not entirely true. I pointed an LLM at the codebase to get me to the right files for understanding skills, and to map out the dependencies and lifecycles - Then I spent quite a bit of time reading the code myself and writing about it.

An AI review at the end of the writing (to "sharpen" the language) unfortunately brought in a couple of AI fingerprints (note the "mic drop" comment above)

edit: write -> right (its 8am)

A technical breakdown of how agent skills are parsed, rendered, injected, and refreshed in your Claude Code working session.

You lost me at "these are formed using radio signals rather than light"