LLMs (Claude Code in particular) will explicitly create token intensive steps, plans and responses - "just to be sure" - "need to check" - "verify no leftovers", will do git diff even tho not asked for, create python scripts for simple tasks, etc.
Absolutely no cache (except the memory which is meh) nor indexing whatsoever.
Pro plan for 20 bucks per month is essentially worthless and, because of this and we are entering new era - the era of $100+ monthly single subscription being something normal and natural.
I'm on the Pro plan. If I run out of tokens, which has only happened 2 or 3 times in months of use, I just work on something else that Claude can't do, or ...write the code myself.
You do have to keep a close eye on it, but I would be doing that anyway given that if it goes haywire it's wasting my time as well as tokens. I'd rather spend an extra minute writing a clearer prompt telling it exactly what I want it to do, than waste time on a slot machine.
>LLMs (Claude Code in particular) will explicitly create token intensive steps, plans and responses - "just to be sure" - "need to check" - "verify no leftovers", will do git diff even tho not asked for, create python scripts for simple tasks, etc. Absolutely no cache (except the memory which is meh) nor indexing whatsoever.
Most of these things can be avoided with a customized CLAUDE.md.
Not in my projects it seems. Perhaps you can share your best practices?
Moreover, avoiding these should be the default behaviour. Currently the default is to drain your pockets.
P.S
CLAUDE.md is sometimes useful but, it's a yet another token drain. Especially that it can grow exponentially.
"A plugin/annotation system where users can teach the agent about custom resource types would scale better than hard-coding each one." - this is a fantastic observation and feedback! Many thanks!
"requiring N consecutive failures before marking down" - I do have the code for it, it's just hidden currently. StatusDude supports 2 types of worker/agents - cloud agents - that will re-verify from multiregion the service status and private agents - the ones we're talking about here - that I might just bring this option back as it makes more sense.
Correlating failures is a bit tricky as usually it requires some sort of manual dependency creation but, I guess for k8s ingress and similar I should be able to figure this out and at least send alerts with appropriate priorities and order.
As for the status page auto generation - currently it's based on namespace - I didn't wanted to bloat the user dashboard too much. Each monitor is tagged with cluster id, namespace and labels. Status Pages pickup monitors based on labels. Users are free to modify these and show exactly what they want :)
It's absolutely stunning that people actually defend this behaviour!
The community is having an outrage - and rightfully so - about a silently discontinued artifact delivery at a very critical time.
Which is their opinion and every human being is entitled to have their own opinion and state it openly.
It is also perfectly fine to expect a standardised behaviour to continue.
However, what is most important is that is perfectly fine to shame an open source product for pulling features and money grabbing people after years of gathering community and locking them in.
I don't think the people in this thread have any concept of how much $$$ it costs to distribute a free container that is going to be downloaded billions of times.
You are a farmer, not a big fancy profitable one. Your tractor is from 1970 and works great, when it works. Your wife has health problems and can't really help out around the farm much - kids have gone off - so you just do things mostly by yourself. With your lucky dog Skip by your side. Even though times are tough and money ain't coming in like it used to - you still give free produce to the local schools and shelters. You've been doing it for over 20 years, and the community loves you for it.
But then your wife passes. Medical bills are too high. You can't give away free produce to the local schools anymore.
The community is outraged. They come to your farm with pitchforks. They set your barn and fields on fire.
> I don't think the people in this thread have any concept of how much $$$ it costs to distribute a free container that is going to be downloaded billions of times.
Not very much at all. It looks like they're hosting on Docker Hub which doesn't charge for bandwidth. I could create a pro account for $11/month and be able to serve an image billions of times. The compute to build an image is small enough that it can be done at whim on a dev machine.
But when you plug in the numbers: that the farmer raised $126 million, and hosting unlimited Docker Hub pulls costs $11/month, it doesn't quite feel the same.
It's more like the farmer was giving leftovers for free to schools and it was so good that it made him famous. People from all over the country came in, including businessmen who told the farmer he is missing out and should be charging more for his food.
He started a restaurant chain but, the businessmen went further and said that a quality product cannot be given away for free and made him stop supporting schools and shelters which got him rich and famous in the first place. Even tho, he was just handing over leftovers (it cost around USD 100 to host a docker image - yearly)
Think EA, Microsoft and Xbox, Broadcom and bitnami.
I don't understand the point. The entire raison d'être of this project is that you self-host it and don't pay money for S3 and control your supply chain.
If you are denied this possibility — it is much easier just to use S3.
There :)
reply