I try to keep agents/Claude.md as tiny as possible. With high level "truths" that don't change. Stack used, invariants, file structure, and some scripts.
Skills are more for things you do often. I run mutation tests, type check,linting,etc. I _could_ just prompt and copy/paste the same prompt each time I need to, or I can just run /tests.
I also have skills for specialized tasks I need every once in a while, like a ux skill, a text skill optimized for xyz, etc.
I think this one might be super interesting for its entertainment value, but there has been some strong criticism as to its credibility. For example, here's a pretty comprehensive debunking from The Skeptical Inquirer: https://skepticalinquirer.org/exclusive/the-telepathy-tapes-...
Framing the ledger as the problem seems strange. It's more accurately the solution. The problem being how you transfer value through the internet in a trusted manner.
Re: privacy though, there's many solutions. Single tailored chains for privacy, mixers, encrypted tokens, permissioned chains, etc.
In my experience, privacy, while important, is not something users actually care about enough to demand a solution for. Most web3 users today just want to degen and gamble, and they're okay with KYC to do so.
Yes - the idea of "not knowing what you don't know" is, in my opinion, one of the most important concepts missing from modern programmers (and probably everybody), especially in the age of LLMs. It's the concept that, personally, most occupies my mind every day, and seems to never come into the minds of many of the innumerable contractors with whom I work.
I have a telegram bot, and I used to have it read message in a group chat to detect the language, and if it wasn't English, it would use gpt5 nano to translate.
To save on inference, I had Claude create a lightweight typescript library to handle translations and detection. Now I get the previous functionality for 200mb~ ram versus paid API credits.
I also built a home assistant menu bar in osx, let's me easily monitor activity, toggle devices, view cameras, etc.
That seems very reasonable, doesn't it? Otherwise the fork could pollute the training data through bad modifications, and tracing that would be a pain.
> every product manager fundamentally misunderstands that users just want a chat window for AI,
I would actually argue chat windows are terrible ui/ux for most cases and users. It does the opposite of `don't make me think`. Too much potential for user error.
Not saying there shouldn't be any LLM integration/features, just that it should be in the form of a button press or something (familiar ux), not the same chatgpt interface that all the early apps are trying to mimic for no good reason.
A tool for non technical people to build things within reason.
I don't know how you're trying to define vibe coding, but I would definitely put it as more of a pejorative/negative term.
I have 20+ years of coding experience, I use coding agents well, and I'm not a vibe coder.
I would also bet that vibe coding will never become good enough to replace me because the models will never have enough context window for an entire repo of a complex app. LLMs can output code for you, but coding is only part of the job.
Not really. Vibecoding is the new code-completion, where every single developer uses extensively but is so mundane that no one bothers to blog about it. You don't see blog posts on how intellisense I'd a kin to magic, or how project starter templates speed up prototyping. No, people just use them and it's so mundane they don't think twice.
Disagree, I think vibe coders should become synonymous with no-code and continue to be somewhat of a pejorative.
I don't like the term vibe engineer, but do agree there needs to be a term to signifiy the difference.
It's also possible in the future just being called a developer/engineer already implies you use coding agents and the people who do it "by hand" will not be the norm.
Skills are more for things you do often. I run mutation tests, type check,linting,etc. I _could_ just prompt and copy/paste the same prompt each time I need to, or I can just run /tests.
I also have skills for specialized tasks I need every once in a while, like a ux skill, a text skill optimized for xyz, etc.
reply