zlacker

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

submitted by hugoda+(OP) on 2026-01-22 18:38:27 | 749 points 635 comments
[view article] [source] [go to bottom]

NOTE: showing posts with links only show all posts
◧◩◪
27. embedd+8b[view] [source] [discussion] 2026-01-22 19:33:35
>>WarmWa+ja
> Anthropic's strategy seems to be to just focus on coding, and they do it well.

Based on their homepage, that doesn't seem to be true at all. Claude Code yes, focuses just on programming, but for "Claude" it seems they're marketing as a general "problem solving" tool, not just for coding. https://claude.com/product/overview

◧◩
51. magicm+Wd[view] [source] [discussion] 2026-01-22 19:46:28
>>areofo+c8
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.

◧◩◪◨
154. furyof+cv[view] [source] [discussion] 2026-01-22 21:26:09
>>kmoser+ui
I mean the comment says they literally don't have support and also complains they don't have a support bot, when they have both.

https://support.claude.com/en/collections/4078531-claude

> As a paid user of Claude or the Console, you have full access to:

> All help documentation

> Fin, our AI support bot

> Further assistance from our Product Support team

> Note: While we don't offer phone or live chat support, our Product Support team will gladly assist you through our support messenger.

158. miohta+Fv[view] [source] 2026-01-22 21:29:17
>>hugoda+(OP)
Luckily there is little vendor lock in and likes of https://opencode.ai/ are picking up the slack
◧◩◪◨⬒
203. cactus+KH[view] [source] [discussion] 2026-01-22 22:43:48
>>throwu+DB
You're right.

https://github.com/anthropics/claude-code/issues

Codex has less but they also had quite a few outages in December. And I don't think Codex is as popular as Claude Code but that could change.

◧◩◪
241. direwo+YV[view] [source] [discussion] 2026-01-23 00:41:01
>>viccis+Gc
I think there's an xkcd alt text about that: https://www.explainxkcd.com/wiki/index.php/1357:_Free_Speech

"I can't remember where I heard this, but someone once said that defending a position by citing free speech is sort of the ultimate concession; you're saying that the most compelling thing you can say for your position is that it's not literally illegal to express."

◧◩◪◨⬒⬓
246. genewi+RW[view] [source] [discussion] 2026-01-23 00:51:23
>>ryandr+4h
https://en.wikipedia.org/wiki/Late_capitalism

https://community.bitwarden.com/t/re-enabling-a-disabled-org...

https://community.meraki.com/t5/Dashboard-Administration/dis...

the former i have heard for a couple decades, the latter is apparently a term of art to prevent hurt feelings or lawsuits or something.

Google thinks i want ADA style organizations, but it's AI caught on that i might not mean organizations for disabled people

btw "ADA" means Americans with Disabilities Act. AI means Artificial Intelligence. A decade is 10 years long. "term of art" is a term of art for describing stuff like jargon or lingo of a trade, skill, profession.

Jargon is specialized, technical language used in a field or area of study. Lingo pins to jargon, but is less technical.

Google is a company that started out crawling the web and making a web search site that they called a search engine. They are now called Alphabet Company (ABC). Crawling means to iteratively parse the characters sent by a webserver and follow links therein, keeping a copy of the text from each such html. HTML is hypertext markup language, hypertext is like text, but more so.

Language is how we communicate.

I can go on?

p.s. if you want a better word, your complaint is about the framing. you didn't gel with the framing of the article. My friend, who holds a doctorate, defended a thesis about how virtually every platform argument is really a framing issue. platform as in, well, anything you care to defend. mac vs linux, wifi vs ethernet, podcasts vs music, guns vs no guns, red vs blue. If you can reduce the frame of the context to something both parties can agree to, you can actually hold a real, intellectual debate, and get at real issues.

248. maz29+yX[view] [source] 2026-01-23 00:58:55
>>hugoda+(OP)
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
◧◩
268. measur+l11[view] [source] [discussion] 2026-01-23 01:32:09
>>tomwph+4p
There are people on twitter bragging about using 100s of agents. Here's one example: https://twitter.com/nearcyan/status/2012948508764946484
◧◩◪◨⬒
290. alexk6+3a1[view] [source] [discussion] 2026-01-23 03:03:53
>>aanet+SU
[BUG] Instantly hitting usage limits with Max subscription: https://github.com/anthropics/claude-code/issues/16157

It's the most commented issue on their GitHub and it's basically ignored by Anthropic. Title mentions Max, but commenters report it for other plans too.

◧◩◪◨
298. riboso+8e1[view] [source] [discussion] 2026-01-23 03:37:42
>>healsd+BF
They did.

>If you want to take a look at the CLAUDE.md that Claude A was making Claude B run with, I commited it and it is available here.

https://github.com/HugoDaniel/boreDOM/blob/9a0802af16f5a1ff1...

327. mohsen+Ko1[view] [source] 2026-01-23 05:28:16
>>hugoda+(OP)
I am doing very similar thing to this and no issue. Even though I am using GLM 4.7 due to cost

I have a complete org hierarchy for Claudes. Director, EM and Worker Claude Code instances working on a very long horizon task.

Code is open source: https://github.com/mohsen1/claude-code-orchestrator

◧◩
384. Markof+yD1[view] [source] [discussion] 2026-01-23 07:45:47
>>Punchy+4C1
Judging by his EUR currency the guy is from EU, so he HAS law available for him to protect himself.

Recital (71) of the GDPR

"The data subject should have the right not to be subject to a decision, which may include a measure, evaluating personal aspects relating to him or her which is based solely on automated processing and which produces legal effects concerning him or her or similarly significantly affects him or her, such as automatic refusal of an online credit application or e-recruiting practices without any human intervention."

https://commission.europa.eu/law/law-topic/data-protection/r...

◧◩
397. Aldipo+HF1[view] [source] [discussion] 2026-01-23 08:05:30
>>dev_l1+uG
Here you are, even open source! And it is a strong one. https://mistral.ai/
◧◩◪
422. mrkeen+sL1[view] [source] [discussion] 2026-01-23 08:57:45
>>chii+ve1
> food

  Can you sell or share farm-saved seed?
  "It is illegal to sell, buy, barter or share farm-saved seed," warns Sam. [1]

  Can feed grain be sown?
  No – it is against the law to use any bought-in grain to establish a crop. [1]

  FTC sues John Deere over farmers' right to repair tractors
  The lawsuit, which Deere called "meritless," accuses the company of withholding access to its technology and best repair tools and of maintaining monopoly power over many repairs. Deere also reaps additional profits from selling parts, the complaint alleges, as authorized dealers tend to sell pricey Deere-branded parts for their repairs rather than generic alternatives. [2]
[1] https://www.fwi.co.uk/arable/the-dos-and-donts-of-farm-saved...

[2] https://www.npr.org/2025/01/15/nx-s1-5260895/john-deere-ftc-...

◧◩◪◨⬒⬓
441. qcnguy+7P1[view] [source] [discussion] 2026-01-23 09:32:09
>>cactus+KH
Don't bother filing issues there. Their issue tracker is a galaxy-sized joke. They automatically close issues after 30 days of inactivity even if they weren't fixed, just to keep the issue count low.

The Reasonable Man might think that an AI company OF ALL COMPANIES would be able to use AI to triage bug tickets and reproduce them, but no! They expect humans to keep wasting their own time reproducing, pinging tickets and correcting Claude when it makes mistakes.

Random example: https://github.com/anthropics/claude-code/issues/12358

First reply from Anthropic: "Found 3 possible duplicate issues: This issue will be automatically closed as a duplicate in 3 days."

User replies, two of the tickets are irrelevant, one didn't help.

Second reply: "This issue has been inactive for 30 days. If the issue is still occurring, please comment to let us know. Otherwise, this issue will be automatically closed in 30 days for housekeeping purposes."

Every ticket I ever filed was auto-closed for inactivity. Complete waste of time. I won't bother filing bugs again.

◧◩◪◨
472. mrkeen+jZ1[view] [source] [discussion] 2026-01-23 10:59:14
>>gregor+gK1
Not a hypothetical

https://lavialibera.it/en-schede-2447-francesca_albanese_und...

◧◩◪
473. Cthulh+xZ1[view] [source] [discussion] 2026-01-23 11:01:01
>>Markof+yD1
More recent, the Digital Services Act includes "Options to appeal to content moderation decisions" [0]; I believe this also covers being banned from a platform. Not sure if Claude falls under these rules, I think it only applies to 'gatekeeper' platforms but I'm reasonably confident the number of organizations that fall under this will increase.

[0] https://digital-strategy.ec.europa.eu/en/policies/digital-se...

501. eibrah+T62[view] [source] 2026-01-23 12:03:33
>>hugoda+(OP)
I did 10k worth of tokens in a month and never had issues with tokens or stuff. I am on the 100 dollar max plan so I did not pay 10k - my wife would have killed me lol

PS: screenshot of my usage (and that was during the holidays https://x.com/eibrahim/status/2006355823002538371?s=46

PPS: I LOVE CLAUDE but I never had to deal with their support so don’t have feedback there

◧◩◪
515. omer_b+Oa2[view] [source] [discussion] 2026-01-23 12:35:21
>>george+Vm1
There is also a big risk for employers' whole organisation to be completely blocked from using Anthropic services if one of their employees have a suspended/banned personal account:

From their Usage Policy: https://www.anthropic.com/legal/aup "Circumvent a ban through the use of a different account, such as the creation of a new account, use of an existing account, or providing access to a person or entity that was previously banned"

If an organisation is large enough and have the means, they MIGHT get help but if the organisation is small, and especially if the organisation is owned by the person whose personal account suspended... then there is no way to get it fixed, if this is how they approach.

I understand that if someone has malicious intentions/actions while using their service they have every right to enforce this rule but what if it was an unfair suspension which the user/employee didn't actually violate any policies, what is the course of action then? What if the employer's own service/product relies on Anthropic API?

Anthropic has to step up. Talking publicly about the risks of AI is nice and all, but as an organisation they should follow what they preach. Their service is "human-like" until it's not, then you are left alone and out.

◧◩◪
523. Purple+pf2[view] [source] [discussion] 2026-01-23 13:07:11
>>urband+PK1
> at what point does a business become so large that it’s not allowed to decide who gets to use it?

It's not about size, it's about justification to fight the ban. You should be able to check if the business has violated your legal rights, or if they even broke their own rules, because failure happens.

> There was a famous case here in the UK of a cake shop that banned a customer for wanting a cake made for a gay wedding because it was contra the owners’ religious beliefs. That was taken all the way up to the Supreme Court IIRC.

I guess it was this one: https://en.wikipedia.org/wiki/Lee_v_Ashers_Baking_Company_Lt...

There was a similar case in USA too: https://en.wikipedia.org/wiki/Masterpiece_Cakeshop_v._Colora...

524. 2sk21+Gf2[view] [source] 2026-01-23 13:08:49
>>hugoda+(OP)
I was reminded of this classic short story by Isaac Asimov, The feeling of Power: https://archive.org/details/1958-02_IF/page/4/mode/2up
◧◩◪◨⬒
537. naaski+qt2[view] [source] [discussion] 2026-01-23 14:29:42
>>kuerbe+k62
Actually, cities often regulate gardens:

https://www.savingadvice.com/articles/2025/07/07/10160132_th...

◧◩◪
550. _joel+gK2[view] [source] [discussion] 2026-01-23 15:52:45
>>mns+7J1
Actually a thing https://www.bbc.co.uk/news/articles/cdd3ep76g3go
552. sambuc+wL2[view] [source] 2026-01-23 15:58:57
>>hugoda+(OP)
If you are in europe you might be able to force them to give you a reason, for an actual human to respond, and who knows maybe even get unbanned.

I have a friend that had a similar experience with amazon, and using an european online platform specific for this he actually got amazon to reopen his business account.

There is a useful list of these european complaints platforms at the bottom of this page: https://digital-strategy.ec.europa.eu/en/policies/dsa-out-co...

◧◩◪◨⬒
565. idonot+CV2[view] [source] [discussion] 2026-01-23 16:45:57
>>starea+aO
Yes, it does exactly that. It also sends other prompts like generating 3 options to choose from, prefilling a reply like 'compile the code', etc. (I can confirm this because I connect CC to llama.cpp and use it with GLM-4.7. I see all these requests/prompts in the llama-server verbose log.)

You can stop most of this with

export DISABLE_NON_ESSENTIAL_MODEL_CALLS=1

And might as well disable telemetry, etc: export CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC=1

I also noticed every time you start CC, it sends off > 10k tokens preparing the different agents. So try not to close / re-open it too often.

source: https://code.claude.com/docs/en/settings

◧◩◪
586. LTL_FT+8p3[view] [source] [discussion] 2026-01-23 19:12:02
>>Camper+UU2
I mean, you could put together a cluster of dgx sparks (8 of them) and hit 100tps with high concurrency:

https://forums.developer.nvidia.com/t/6x-spark-setup/354399/...

Or a single user at about 10tps.

This is probably around $30k if you go with the 1tb models.

◧◩◪◨
614. deaux+Aq4[view] [source] [discussion] 2026-01-24 02:07:11
>>daft_p+HI2
Makes sense, I agree for general chat Gemini is great.

For Gemini Code Assist [1], the problem remains that their models are very poor at tool calling and that their harness (Gemini CLI) is miles behind. Looks like there's a plugin for Opencode to use this subscription, which helps with the harness part.

If Gemini 3 GA - which is taking suspiciously long - is better at tool calling then it'll be a great option.

[1] https://cloud.google.com/products/gemini/pricing

◧◩◪◨⬒⬓
615. DANmod+7z4[view] [source] [discussion] 2026-01-24 04:02:11
>>sigseg+oc1
13 days ago on HN:

>>46566292

◧◩◪◨⬒⬓
629. thadk+Mr8[view] [source] [discussion] 2026-01-25 17:11:00
>>vbezhe+831
this Nov video talks thru via Cursor renegotiation, how in important ways a token != a token (not even within types like reasoning, output, or a model) because attention can lead to expensive logic loops. https://www.oreilly.com/videos/live-with-tim/0642572259488/ https://www.youtube.com/watch?v=6AXMO2dW0LI&list=PL055Epbe6d...
[go to top]