Hacker Newsnew | past | comments | ask | show | jobs | submit | strictnein's commentslogin

You can use OpenCode programmatically, thus turning that $200/mo Claude Code account into a very cheap Opus 4.6 API service.

I don't think there's anything really to it past that.


you can also use Claude Code programmatically with the `prompt` parameter in the cli.

And they can ban your account if they think you are doing that. I think someone even commented here on HN they were banned by Anthropic for this.

Why would they have that feature in claude code cli if it goes against the ToS? You can use Claude Code programatically. This is not the issue. The issue is that Anthropic wants to lock you in within their dev ecosystem (like Apple does). Simple as that.

allowed shell pipes doesn't necessarily mean they want loops running them.

One of the economic tuning features of an LLM is to nudge the LLM into reaching conclusions and spending the tokens you want it to spend for the question.

presumably everyone running a form of ralph loop against every single workload is a doomsday situation for LLM providers.


> allowed shell pipes doesn't necessarily mean they want loops running them.

insane that people apologize for this at all. we went from FOSS software being standard to a proprietary cli/tui using proprietary models behind a subscription. how quickly we give our freedom away.


I don't know why this is downvoted, see my nephew (?) comment [0] for a longer version, but this is not at all clear IMHO. I'm not sure if a "claude -p" on a cron is allowed or not with my subscription, if I run it on another server is it? Can I parse the output of claude (JSON) and have another "claude -p" instance work on the response? It's only a hop, skip, and a jump over to OpenClaw it seems, which is _not_ allowed. But at what point did we cross the line?

It feels like the only safe thing to do is use Claude Code, which, thankfully, I find tolerable, but unfortunate.

[0] https://news.ycombinator.com/item?id=47446703


is this against their tos or something? what did they expect programmers to do, not automate the automated code-writer?

Or the Python or Typescript Agent SDK libraries: https://platform.claude.com/docs/en/agent-sdk/overview

Or can you? It's my understanding that you cannot use your subscription with the Agent SDK, that's what the docs say:

> Unless previously approved, Anthropic does not allow third party developers to offer claude.ai login or rate limits for their products, including agents built on the Claude Agent SDK. Please use the API key authentication methods described in this document instead.

Though there was that tweet [0] a while back by someone from Anthropic that just muddied the water. It's frustrating because I feel like the line between the Agent SDK and `claude -p` is not that large but one can use the subscription and one can't... or we don't know, the docs seem unambiguous but the tweet confuses things and you can find many people online saying you can, or you can't.

I'd love to play around with the Agent SDK and try out some automations but it seems I can only do that if I'm willing to pay for tokens, even though I could use Claude Code to write the code "for" the Agent SDK, but not "run" the Agent SDK.

Where is the line? Agent SDK is not allowed with subscription, but if I write a harness around passing data to and parsing the JSON response from `claude -p '<Your Prompt>' --output-format json` would that be allowed? If I run it on a cron locally? I literally have no idea and, not wanting my account to be banned, I'm not interested in finding out. I wish they would clarify it.

[0]

Twitter: https://x.com/trq212/status/2024212378402095389

XCancel: https://xcancel.com/trq212/status/2024212378402095389

Text:

> Apologies, this was a docs clean up we rolled out that’s caused some confusion.

> Nothing is changing about how you can use the Agent SDK and MAX subscriptions!

> We want to encourage local development and experimentation with the Agent SDK and claude -p.

> If you’re building a business on top of the Agent SDK, you should use an API key instead. We’ll make sure that’s clearer in our docs.


Yea I am this exact boat - it's insane how unclear it is. Why have the -p option at all if they don't want it to be called by other processes?

If they only want people using claude code inside their harness, they could... just remove -p?


> You can use OpenCode programmatically, thus turning that $200/mo Claude Code account into a very cheap Opus 4.6 API service.

Can you explain what you mean by this?


You can easily automate OpenCode - more so than the basic Claude Code or Claude desktop app - in a way that automatically uses the maximum amount of subscription quota every cycle. And in an inefficient way that Anthropic can't cache on their end.

If you know anything about subscription models, you know that ALL of them are built on the fact that most of the users don't use the full capacity available all the time.


Their SDK shows a really basic example that you could build out pretty easily, ironically about Anthropic:

https://opencode.ai/docs/sdk/#structured-output

You can stand up an OpenAI compatible API layer in front of it and just feed the requests back and forth. Adds a little delay, but not much.


Can you show any service that is selling API because they have turned a $200/mo Claude code account into an API service? Give me a break.

I mean, first and foremost, Tiktok has offices in the US and employees thousands of people here.

Unless things are really different in California from where I'm at, I find it unlikely that the city is responsible for maintaining state and federal highways.


At least in my state, state and federal highways are all the responsibility of the state DOT, which is massively subsidized out of general fund tax dollars and federal grants. There have been recent attempts to increase fuel taxes, but they get regularly shut down by the electorate. It's not looking good for highway maintenance, either.


> The amount of breathless conclusion jumping from citizen journalists has been completely bananas as well.

I saw a video on Instagram claiming (or at least insinuating) that Epstein had access to all of Lifetouch's photos, because "the company is in the Epstein files!". Turns out, it was a single line item for $109 in what looked like banking records. In comparison, that same selection of files mentioned Whole Foods ~250 times.

For those outside the US, Lifetouch does school photos for about half the schools in the US (or something like that) so you can understand how that's a thread that conspiracy theorists can pull on. But there's nothing there. Just a single payment to the company in a sea of thousands and thousands of normal, every day purchases.


[flagged]


There is a book “Nothing Is True and Everything Is Possible” which you might want to consider before deciding truth doesn’t matter anymore. It describes an explicit strategy by the Kremlin to poison the information landscape with lies, half truths, and conspiracies. And amplify conflicting narratives.

Eventually people stop caring about what is true anymore.


> It describes an explicit strategy by the Kremlin to poison the information landscape with lies, half truths, and conspiracies. And amplify conflicting narratives.

For someone who wants an intro to the subject, this 2016 paper by RAND is pretty good; 'The Russian "Firehose of Falsehood" Propaganda Model':

> The experimental psychology literature suggests that, all other things being equal, messages received in greater volume and from more sources will be more persuasive. Quantity does indeed have a quality all its own. High volume can deliver other benefits that are relevant in the Russian propaganda context. First, high volume can consume the attention and other available bandwidth of potential audiences, drowning out competing messages. Second, high volume can overwhelm competing messages in a flood of disagreement. Third, multiple channels increase the chances that target audiences are exposed to the message. Fourth, receiving a message via multiple modes and from multiple sources increases the message's perceived credibility, especially if a disseminating source is one with which an audience member identifies.

* https://www.rand.org/pubs/perspectives/PE198.html

See also Steve Bannon's 'flood the zone' technique:

* https://en.wikipedia.org/wiki/Flood_the_zone

Bannon seems to have connections with Epstein (who is may have links to Russia):

* https://en.wikipedia.org/wiki/Steve_Bannon#Connection_to_Jef...


“Flood the zone” is the speech version of what the Kremlin does.

If it came out that some anti-Trump terrorists were actually backed by MAGA, it would be analogous to what the Kremlin does.


> Eventually people stop caring about what is true anymore.

In the age of easy to fake video, photos, audio, etc, who wouldn't be?


nixosbestos’s comment was in response to a post about how an arbitrary brand new conspiracy theory didn’t have compelling evidence in the emails. It seems like saying “who cares” to what kind of reads like “technically there’s an infinite amount of conspiracy theories that aren’t proven in the emails” is saying the opposite of “truth doesn’t matter”


Fuck the Kremlin. Corporate America calls the same damn thing marketing, advertising, sales, and public relations respectively.


Kind of. Corporate propaganda (ads) are very much still in the old Bush-era paradigm of “repeating a lie until it becomes true”. Coke and Pepsi don’t want you to believe “nothing is true”. They want you to believe their cola is superior. Coke isn’t going to waste money creating confusion about that topic.

The Kremlin will amplify conflicting opinions. Even critics of the government. Later it will come out that those critics were sponsored by Putin, undermining any critics by association.


Thank you.


Except you and everyone you agree with I'm sure


In the second half of my 40s now and I'm in the same boat. I started slapping keys on a c64 when I was 2 years old. Really enjoyed software development until 10-15 years ago. With the current LLM tooling available the number of systems I've been able to build that are novel and tackle significant problems has been kind of mind blowing over the past 8 months or so.

Staying up late, hacking away at stuff like I used to, and it's been a blast.

Finally, Homeworld was awesome and it felt magical playing it.


I think there's still quite a chasm out there. Domain knowledge, an informed and opinionated view on how something should function, and overall tech knowledge are still key. Having those three things continues to greatly differentiate people of equal coding skill, as they always have.


That’s something LLMs are also presumably good at. At least I’m seeing more and more push to use LLMs at work for ambitious business requirements instead of learning about the problem we’ve been dealing with. Instead of knowing why you are doing what you’re doing, now people are just asking LLMs for specific answers and move on.

Sure some might use it to learn as well, but it’s not necessary and people just yolo the first answer claude gives to them.


Yeah, but I used to be a wizard with arcane knowledge making computers do things others didn't even understand. I was casting fireballs, and now everyone has Find Greater Familiar right out of the gate who does all the heavy lifting. :(


> "tfile kids"

Not familiar with that term, and my googling has failed. What does it refer to?


http://textfiles.com/

people familiar with the culture


Works much better than I thought it would. It's rather rare when I see an add in Chrome.


Plenty of states have eliminated exhaust inspections. They were wholly ineffective and barely "caught" anyone.


Huh? The US has the largest private investment pool in the world.

Why would investors invest their money in things that have no chance of recouping that investment?


Exactly! It's an absolutely foolish thing to build a society around, and the benefits are largely squandered on the private lives of private investors.


"Absolutely foolish" - the tech sector doesn't exist without this "foolishness", but other than that, great idea.


Is that necessarily a bad thing? There are other ways to develop technology.


Nd they’re all concerned with next quarters results, not the next hundred years.


Are your investments in places where you're interested in the next 100 years and won't see the benefits until after your dead?


That belies why relying on individual investors to guide society is never going to work. You might as well cede the globe over to whichever society had the balls to centralize the economy


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: