This brings me back to my college days. We had Windows, and Deep Freeze. Students could do anything on the computer, we restart it and its all wiped and new. How long before Deep Freeze realizes they could sell their tool to Vibe Coders, they have Deep Freeze for Mac but not for Linux, funnily enough.
If that's the goal, why not just have Claude Code do it all from your phone at that point? Test it when its done locally you pull down the branch. Not 100% frictionless, but if it messes up an OS it would be anthropic's not yours.
I believe Perplexity is doing this already, but specifically for looking up products, which is how I use AI sometimes. I am wondering how long before eBay, Amazon etc partner with AI companies to give them more direct API access so they can show suggested products and what not. I like how AI can summarize things for me when looking up products, then I open up the page and confirm for myself.
The best analogy I think is, if you just take Stack Overflow code solutions, smoosh over your code and hit compile / build, and move on without ever looking at "why it works" you're really not using your skills to the best of your ability, and it could introduce bugs you didn't expect, or completely unnecessary dependencies. With Stack Overflow you can have other people pointing out the issues with the accepted answer and giving you better options.
This keeps coming up again and again and again, but like how many times were you able to copy paste SO solution wholesale and just have it work? Other than for THE most simple cases (usually CSS) there would always have to be some understanding involved. Of course you don't always learn deeply every time, but the whole "copy paste off of stackoverflow" was always an exaggeration that is being used in seeming earnest.
> This looks like plain political pressure. No lives were saved, and no crime was prevented by harassing local workers.
I wouldn't even consider this a reason if it wasn't for the fact that OpenAI and Google, and hell literally every image model out there all have the same "this guy edited this underage girls face into a bikini" problem (this was the most public example I've heard so I'm going with that as my example). People still jailbreak chatgpt, and they've poured how much money into that?
Claude reads from .claude/instructions.md whenever you make a new convo as a default thing. I usually have Claude add things like project layout info and summaries, preferred tooling to use, etc. So there's a reasonable expectation of how it should run. If it starts 'forgetting' I tell it to re-read it.
No, Claude Code reads the CLAUDE.md in the root of your project. It's case sensitive so it has to be exactly that, too. Github Copilot reads from .github/copilot-instructions.md and supposedly AGENTS.md. Anigravity reads AGENTS.md and pulls subagents and the like from a .agents directory. This is probably why you have to remind it to re-read it so much, the harness isn't loading it for you.
My annoyance is that it sounds like I can't just use Claude Code directly in XCode? I like how Zed does it, it's not perfect, but it works really nicely.
Xcode is using the Claude Agent SDK, which means that you "get the full power of Claude Code directly in Xcode—including subagents, background tasks, and plugins—all without leaving the IDE¹". I assume that means iOS development plug-ins like Axiom² should work as well.
I buy the theory that Claude Code is engineered to use things like token caching efficiently, and their Claude Max plans were designed with those optimizations in mind.
If people start using the Claude Max plans with other agent harnesses that don't use the same kinds of optimizations the economics may no longer have worked out.
(But I also buy that they're going for horizontal control of the stack here and banning other agent harnesses was a competitive move to support that.)
It should just burn quota faster then. Instead of blocking they should just mention that if you use other tools then your quota may reduce at 3x speed compared to cc. People would switch.
When I last checked a few months ago, Anthropic was the only provider that didn't have automatic prompt caching. You had to do it manually (and you could only set checkpoints a few times per context?), and most 3rd party stuff does not.
They seem to have started rejecting 3rd party usage of the sub a few weeks ago, before Claw blew up.
By the way, does anyone know about the Agents SDK? Apparently you can use it with an auth token, is anyone doing that? Or is it likely to get your account in trouble as well?
Absolutely. I installed clawdbot for just long enough to send a single message, and it burned through almost a quarter of my session allowance. That was enough for me. Meanwhile I can use CC comfortably for a few hours and I've only hit my token limit a few times.
I've had a similar experience with opencode, but I find that works better with my local models anyway.
I have a feeling the different harnesses create new context windows instead of using one. The more context windows you open up with Claude the quicker your usage goes poof.
I would be surprised if the primary reason for banning third party clients isn't because they are collecting training data via telemetry and analytics in CC. I know CC needlessly connects to google infrastructure, I assume for analytics.
In what way would it be abused? The usage limits apply all the same, they aren't client side, and hitting that limit is within the terms of the agreement with Anthropic.
The subscription services have assumptions baked in about the usage patterns; they're oversubscribed and subsidized. If 100% of subscriber customers use 100% of their tokens 100% of the time, their business model breaks. That's what wholesale / API tokens are for.
> hitting that limit is within the terms of the agreement with Anthropic
It's not, because the agreement says you can only use CC.
This is how every cloud service and every internet provider works. If you want to get really edgy you could also say it's how modern banking works.
Without knowing the numbers it's hard to tell if the business model for these AI providers actually works, and I suspect it probably doesn't at the moment, but selling an oversubscribed product with baked in usage assumptions is a functional business model in a lot of spaces (for varying definitions of functional, I suppose). I'm surprised this is so surprising to people.
Don't forget gyms and other physical-space subscriptions. It's right up there with razor-and-blades for bog standard business models. Imagine if you got a gym membership and then were surprised when they cancelled your account for reselling gym access to your friends.
If they rely on this to be competitive, I have serious doubts they will survive much longer.
There are already many serious concerns about sharing code and information with 3rd parties, and those Chinese open models are dangerously close to destroying their entire value proposition.
> selling an oversubscribed product with baked in usage assumptions is a functional business model in a lot of spaces
Being a common business model and it being functional are two different things. I agree they are prevalent, but they are actively user hostile in nature. You are essentially saying that if people use your product at the advertised limit, then you will punish them. I get why the business does it, but it is an adversarial business model.
>Without knowing the numbers it's hard to tell if the business model for these AI providers actually works
It'll be interesting to see what OpenAI and Anthropic will tell us about this when they go public (seems likely late this year--along with SpaceX, possibly)
The Business model is Uber. It doesn't work unless you corner the market and provide a distinct value replacement.
The problem is, there's not a clear every-man value like Uber has. The stories I see of people finding value are sparse and seem from the POV of either technosexuals or already strong developer whales leveraging the bootstrapy power .
If AI was seriously providing value, orgs like Microsoft wouldn't be pushing out versions of windows that can't restart.
It clearly is a niche product unlike Uber, but it's definitely being invested in like it is universal product.
> Selling dollars for $.50 does that. It sounds like they have a business model issue to me.
its not. The idea is that majority subscribers don't hit limit, so they sell them dollar for 2. But there is minority which hit limit, and they effectively selling them dollar for 50c, but aggregated numbers could be positive.
> It's not, because the agreement says you can only use CC.
it's like Apple: you can use macOS only on our Macs, iOS only on iPhones, etc. but at least in the case of Apple, you pay (mostly) for the hardware while the software it comes with is "free" (as in free beer).
Taking umbrage as if it matters how I use the compute I'm paying for via the harness they want me to use it within as long as I'm just doing personal tasks I want to do for myself, not trying to power an apps API with it seems such a waste of their time to be focusing on and only causes brand perception damage with their customers.
The loss of access shows the kind of power they'll have in the future. It's just a taste of what's to come.
If a company is going to automate our jobs, we shouldn't be giving them money and data to do so. They're using us to put ourselves out of work, and they're not giving us the keys.
I'm fine with non-local, open weights models. Not everything has to run on a local GPU, but it has to be something we can own.
I'd like a large, non-local Qwen3-Coder that I can launch in a RunPod or similar instance. I think on-demand non-local cloud compute can serve as a middle ground.
How do I "abuse" a token? I pass it to their API, the request executes, a response is returned, I get billed for it. That should be the end of the conversation.
(Edit due to rate-limiting: I see, thanks -- I wasn't aware there was more than one token type.)
> BMW actually has a reasonable control surface though, not a grand user interface experiment by some crack heads.
Really? It's one thing to hate Elon Musk, but you're talking about a lot of brilliant engineers who worked on these cars, everything from the components to the software. It's uneeded low blow just because you don't like Elon Musk.
The UX is a mess. Why does the car always label the trunk as open rather than have a button that I press to open it?
Why does cruise control sometimes change to the speed limit and sometimes not?
Why does auto lane change sometimes need me to start the manoeuvre and sometimes not? If I guess wrong and start the lane change myself, all autopilot just disengages suddenly.
I have to proove that I'm holding the wheel by wiggling it from time to time, but if I accidentally wiggle too hard it disengages. Why not have a sensor or use the cameras to detect if I'm holding the wheel?
My son didn't shut the back door properly. I started driving and the car started binging. It didn't tell me why it was binging until I put it in park and looked at the pretty 3d representation of the car, then noticed that the door was open.
Maybe if I drove more regularly I would get used to all this stuff. The car was borrowed and I gave it back.
I’m glad you found a place to get these complaints off your chest, but these are kind of hilarious. the button says “open trunk”. It’s a verb. If this is your complaint then lmao have you not seen what other OEM software looks like? Door open doesn’t just ding, it shows a warning with plain english explanation and an icon.
For the rest of your complaints you can mostly thank the overzealous EU/unece regulation which limits steering torque and requires intervention. FSD has none of those concerns, it just drives and does not require torque on the wheel.
reply