I was banned from Claude for scaffolding a Claude.md file?

I've been using Claude Code with AWS Bedrock as the provider. Setup guide if you're interested: https://code.claude.com/docs/en/amazon-bedrock

> Or I don't know. This is all just a guess from me.

OT: Has anyone observed that Claude Code in CLI works more reliably than the web or desktop apps?

I can run very long, stable sessions via Claude Code, but the desktop app regularly throws errors or simply stops the conversation. A few weeks ago, Anthropic introduced conversation compaction in the Claude web app. That change was very welcome, but it no longer seems to work reliably. Conversations now often stop progressing. Sometimes I get a red error message, sometimes nothing at all. The prompt just cannot be submitted anymore.

I am an early Claude user and subscribed to the Max plan when it launched. I like their models and overall direction, but reliability has clearly degraded in recent weeks.

Another observation: ChatGPT Pro tends to give much more senior and balanced responses when evaluating non-technical situations. Claude, in comparison, sometimes produces suggestions that feel irrational or emotionally driven. At this point, I mostly use Claude for coding tasks, but not for project or decision-related work, where the responses often lack sufficient depth.

Lastly, I really like Claude’s output formatting. The Markdown is consistently clean and well structured, and better than any competitor I have used. I strongly dislike ChatGPT’s formatting and often feed its responses into Claude Haiku just to reformat them into proper Markdown.

Curious whether others are seeing the same behavior.

is there a benefit of using a separate claude instance to update the CLAUDE.md of the first? I always want to leverage the full context of the situation to help describe what went wrong, so doing it "inline" makes more sense.

We need local models asap.

It should be mentionned in the title that these are just speculations.

Is it time to move to open source and run model locally with an DGX Spark?

I was banned from just trying out Claude AI chat for the first time a few months ago. I emailed them and restored my account access.

do you know for sure this was the reason why?

RIP. I hear they're looking for janitors.

Claude is going wild lately. It told me I had used up 75% of my weekly limit. Ohhhk. I sent one more short query, and boom blocked til til Monday because i used up 25% in that one go (on thursday). How is that possible? Its falling off fast right now.

There needs to be a law that prevents companies from simply banning you, especially when it's an important company. There should be an explanation and they shouldn't be allowed to hide behind some veil. There should be a real process with real humans that allow for appeals etc instead of scripts and bots and automated replies.

Thinking 220GBP for a high-limit Claude account is the kind of thinking that really takes for granted the amount of compute power being used by these services. That's WITH the "spending other people's money" discount that most new companies start folks off with. The fact that so many are painfully ignorant of the true externalities of these technologies and their real price never ceases to amaze me.

Not that it’s the same thing, but how real is it to have a locally setup model for coding?

Granted, it’s not going to be Claude scale but it’d be nice to do some of it locally.

That's why I run a local Qwen3-Next model on an NVIDIA Thor dev kit (Apple Silicon and DGX Spark are other options but they are even more expensive for 128GB VRAM)

Another instance of "Risk Department Maoism".

If you're wondering, the "risk department" means people in an organization who are responsible for finding and firing customers who are either engaged in illegal behavior, scamming the business, or both. They're like mall rent-a-cops, in that they don't have any real power beyond kicking you out, and they don't have any investigatory powers either. But this lack of power also means the only effective enforcement strategy is summary judgment, at scale with no legal recourse. And the rules have to be secret, with inconsistent enforcement, to make honest customers second-guess themselves into doing something risky. "You know what you did."

Of course, the flipside of this is that we have no idea what the fuck Hugo Daniel was actually doing. Anthropic knows more than we do, in fact: they at least have the Claude.md files he was generating and the prompts used to generate them. It's entirely possible that these prompts were about how to write malware or something else equally illegal. Or, alternatively, Anthropic's risk department is just a handful of log analysis tools running on autopilot that gave no consideration to what was in this guy's prompts and just banned him for the behavior he thinks he was banned for.

Because the risk department is an unaccountable secret police, the only recourse for their actions is to make hay in the media. But that's not scalable. There isn't enough space in the newspaper for everyone who gets banned to complain about it, no matter how egregious their case is. So we get all these vague blog posts about getting banned for seemingly innocuous behavior that could actually be fraud.

This is very cool. I looked at the Claude.md he was generating and it is basically all of Claude's failure modes in one file. I can think of a few reasons why Anthropic would not want this information out in the open or for someone to systematically collate all the data into one file.

In Open WebUI I have different system prompts (startup advisor, marketing expert, expert software engineer etc) defined and I use Claude via OpenRouter.

Is this going to get me banned? If so i'll switch to a different non-anthropic model.

Scamthropic at it again

Why are so many people so obsessed with feeding as many prompts/data as possible to LLMs and generating millions of lines of code?

What are you gonna do with the results that are usually slop?

Hmm so how are the alternatives? Just in case I will get banned for nothing as well. I’m riding cc with opus all day long these days.

> Like a lot of my peers I was using claude code CLI regularly and trying to understand how far I could go with it on my personal projects. Going wild, with ideas and approaches to code I can now try and validate at a very fast pace. Run it inside tmux and let it do the work while I went on to do something else

This blog post could have been a tweet.

I'm so so so tired of reading this style of writing.

bow down to our new overlords - dont' like it? banned, with no recourse - enjoy getting left behind, welcome to the future old man

Well at least they didn't email the press and called the FBI on you?

> I got my €220 back (ouch that's a lot of money for this kind of service, thanks capitalism).

I'm not sure I understand the jab here at capitalism. If you don't want to pay that, then don't.

Isn't that the point of capitalism?

[dead]

[dead]

[dead]

Just stop using Anthropic. Claude Code is crap because they keep putting in dumb limits for Opus.

I always take these sorts of "oh no I was banned while doing something innocent" posts with a large helping of salt. At least the ones where someone is complaining about a ban from Stripe, usually it turns out they are doing something that either violates the terms of service or is actually fraudulent. None the less its quite frustrating dealing with these because either way.

You mean the throwaway pseudonym you signed up with was banned, right?

… right ?

The news is not that they turned off this account. The news is that this user understands very little about the nature of zero sum context mathematics. The mentioned Claude.md is a totally useless mess. Anthropic is just saving themselves from the token waste of this strategy on a fixed billing rate plan.

If the OP really wants to waste tokens like this, they should use a metered API so they are the one paying for the ineffectiveness, not Anthropic.

(Posted by someone who has Claude Max and yet also uses $1500+ a month of metered rate Claude in Kilo Code)

This feels... reasonable? You're in their shop (Opus 4.5) and they can kick you out without cause.

But Claude Code (the app) will work with a self-hosted open source model and a compatible gateway. I'd just move to doing that.