I was banned from Claude for scaffolding a Claude.md file? • Hacker News Board

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

#51

Post by munk-a » Thu, Jan 22, 2026, 7:43 PM UTC

areoform wrote: post #11

I recently found out that there's no such thing as Anthropic support. And that made me sad, but not for reasons that you expect. Out of all of the tech organizations, frontier labs are the one org you'd expect to be trying out cutting edge forms of support. Out of all of the different things these agents can do, surely most forms of "routine" customer support are the lowest hanging fruit? I think it's possible for An…

> They're begging corporate decision makers to ask the question, "If Anthropic doesn't trust Claude to run its support, then why should we?"

Don't worry - I'm sure they won't and those stakeholders will feel confident in their enlightened decision to send their most frustrated customers through a chatbot that repeatedly asks them for detailed and irrelevant information and won't let them proceed to any other support levels until it is provided.

I, for one, welcome our new helpful overlords that have very reasonably asked me for my highschool transcript and a ten page paper on why I think the bug happened before letting me talk to a real person. That's efficiency.

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

#52

Post by 0xbadcafebee » Thu, Jan 22, 2026, 7:45 PM UTC

WarmWash wrote: post #25

areoform wrote: post #11

I recently found out that there's no such thing as Anthropic support. And that made me sad, but not for reasons that you expect. Out of all of the tech organizations, frontier labs are the one org you'd expect to be trying out cutting edge forms of support. Out of all of the different things these agents can do, surely most forms of "routine" customer support are the lowest hanging fruit? I think it's possible for An…

Claude is an amazing coding model, its other abilities are middling. Anthropic's strategy seems to be to just focus on coding, and they do it well.

Critically, this has to be their play, because there are several other big players in the "commodity LLM" space. They need to find a niche or there is no reason to stick with them.

OpenAI has been chaotically trying to pivot to more diversified products and revenue sources, and hasn't focused a ton on code/DevEx. This is a huge gap for Anthropic to exploit. But there are still competitors. So they have to provide a better experience, better product. They need to make people want to use them over others.

Famously people hate Google because of their lack of support and impersonality. And OpenAI also seems to be very impersonal; there's no way to track bugs you report in ChatGPT, no tickets, you have no idea if the pain you're feeling is being worked on. Anthropic can easily make themselves stand out from Gemini and ChatGPT by just being more human.

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

#53

Post by swiftcoder » Thu, Jan 22, 2026, 7:45 PM UTC

landryraccoon wrote: post #38

This blog post feels really fishy to me. It's quite light on specifics. It should have been straightforward for the author to excerpt some of the prompts he was submitting, to show how innocent they are. For all I know, the author was asking Claude for instructions on extremely sketchy activity. We only have his word that he was being honest and innocent.

> It should have been straightforward for the author to excerpt some of the prompts he was submitting

If you read to the end of the article, he links the committed file that generates the CLAUDE.md in question.

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

#54

Post by WarmWash » Thu, Jan 22, 2026, 7:45 PM UTC

arcanemachiner wrote: post #48

WarmWash wrote: post #25

Earlier quoted context omitted.

Claude is an amazing coding model, its other abilities are middling. Anthropic's strategy seems to be to just focus on coding, and they do it well.

Interesting. Would anyone care to chime in with their opinion of the best all-rounder model?

You'll get 30 different opinions and all those will disagree with each other.

Use the top models and see what works for you.

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

#55

Post by andrelaszlo » Thu, Jan 22, 2026, 7:46 PM UTC

alistairSH wrote: post #16

tobyhinloopen wrote: post #7

So you were generating and evaluating the performance of your CLAUDE.md files? And you got banned for it?

It reads like he had a circular prompt process running, where multiple instances of Claude were solving problems, feeding results to each other, and possibly updating each other's control files?

Could anyone explain to me what the problem is with this? I thought I was fairly up to date on these things, but this was a surprise to me. I see the sibling comment getting downvoted but I promise I'm asking this in good faith, even if it might seem like a silly question (?) for some reason.

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

#56

Post by magicmicah85 » Thu, Jan 22, 2026, 7:46 PM UTC

areoform wrote: post #11

I recently found out that there's no such thing as Anthropic support. And that made me sad, but not for reasons that you expect. Out of all of the tech organizations, frontier labs are the one org you'd expect to be trying out cutting edge forms of support. Out of all of the different things these agents can do, surely most forms of "routine" customer support are the lowest hanging fruit? I think it's possible for An…

https://support.claude.com/en/articles/9015913-how-to-get-su...

Their support includes talking to Fin, their AI support with escalations to humans as needed. I dont use Claude and have never used the support bot, but their docs say they have support.

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

#57

Post by NitpickLawyer » Thu, Jan 22, 2026, 7:46 PM UTC

epolanski wrote: post #22

alistairSH wrote: post #16

Earlier quoted context omitted.

It reads like he had a circular prompt process running, where multiple instances of Claude were solving problems, feeding results to each other, and possibly updating each other's control files?

What would be bad in that? Writing the best possible specs for these agents seems the most productive goal they could achieve.

I think the idea is fine, but what might end up happening is that one agent gets unhinged and "asks" another agent to do more and more crazy stuff, and they get in a loop where everything gets flagged. Remember that "bots configured to add a book at +0.01$ on amazon, reached 1M$ for the book" a while ago. Kinda like that, but with prompts.

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

#58

Post by munk-a » Thu, Jan 22, 2026, 7:47 PM UTC

preinheimer wrote: post #3

> AI moderation is currently a "black box" that prioritizes safety over accuracy to an extreme degree. I think there's a wide spread in how that's implemented. I would certainly not describe Grok as a tool that's prioritized safety at all.

You say that - and yet it has successfully guarded Elon from any of those pesky truths that might harm his fervently held beliefs. You just forgot to consider that Grok is a tool that prioritizes Elon's emotional safety over all other safeties.

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

#59

Post by staticman2 » Thu, Jan 22, 2026, 7:48 PM UTC

lazyfanatic42 wrote: post #29

alistairSH wrote: post #12

Earlier quoted context omitted.

You're not alone. I think the author was doing some sort of circular prompt injection between two instances of Claude? The author claims "I'm just scaffolding a project" but that doesn't appear to be the case, or what resulted in the ban...

Author really comes off unhinged throughout the article to be frank.

Author thinks he's cute to do things like mention Google without typing Google but I wouldn't call him unhinged.

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

#60

Post by raincole » Thu, Jan 22, 2026, 7:48 PM UTC

superb_dev wrote: post #31

cortesoft wrote: post #5

I am really confused as to what happened here. The use of ‘disabled organization’ to refer to the author made it extra confusing. I think I kind of have an idea what the author was doing, but not really.

The author was using instance A of Claude to update a `claude.md` while another instance B of Claude was consuming that file. When Claude B did something wrong, the author asked Claude A to update the `claude.md` so that Claude B didn’t make the same mistake again

Which shouldn't be bannable imo. Rate throttle is a more reasonable response. But Anthropic didn't reply to the author, so we don't even know if it's the real reason they got banned.