I think the suspicion is based on this app being offered in a region whose government is hostile to privacy and this implementation being connected with the strong nativist bent in Europe.

The "spec" is not relevant in any way because we have no idea what else is going on. Why was it relevant that these operators must specifically be in the EU? Everyone is just complying with the global spec...but the app provider must be in Europe...okay.

> Why was it relevant that these operators must specifically be in the EU

The integration is only possible because the EU forced Meta's hand. The law only applies to massive digital empires with gatekeeper levels of control.

I don't think the EU would mind at all if Meta would permit American companies to interoperate with them. Meta won't just permit it, they have to protect their WhatsApp Business money machine of course.

That's also why the feature is only available to EU numbers. Not because BirdyChat hates Australians, but because WhatsApp won't permit them to send messages to numbers from those countries.

> region whose government is hostile to privacy

Which government?

EU. I don't think it is any better at the national level however.

The EU is not a government. It's a loose economic confederation. And national European governments vary wildly in their positions on this.

It isn't an "economic confederation". It has a parliament, an executive, a judiciary, and a civil service. I would read the wiki page on the European Union.

The EU parliament can't propose laws, unlike any parliament in the world.

The executive is formed out of national government heads of state, which can veto everything.

Its judiciary and actually all 3 branches are strictly limited in their powers to powers delegated to them (which are weaker than the US Articles of Confederation).

The civil service is covered by the comments above.

In technical terms it is a government, in real life is is strictly limited, albeit growing. No country could operate with the "government" the EU has. France has several million government employees for about 70 million people while the EU has at most 50 000 workers for 450 million citizens).

This is a very complicated topic and I don't really apreciate the condescension inherent in sending me to Wikipedia.

Call it what you want but the fact remains that they can write a lot of laws the member countries must follow, for better or worse. GDPR, Chat Control, etc.

It isn't sub agents. The gap with existing tooling is that the abstraction is over a task rather than a conversation (due to the issue with third-party apps, Claude Code has been inherently limited to conversations which is why they have been lacking in this area, Claude Code Web was the first move in this direction), and the AI is actually coordinating the work (as opposed to being constantly prompted by the user).

One of the issues that people had which necessitated this feature is that you have a task, you tell Claude to work on it, and Claude has to keep checking back in for various (usually trivial) things. This workflow allows for more effective independent work without context management issues (if you have subagents, there is also an issue with how the progress of the task is communicated by introducing things like task board, it is possible to manage this state outside of context). The flow is quite complex and requires a lot of additional context that isn't required with chat-based flow, but is a much better way to do things.

The way to think about this pattern - one which many people began concurrently building in the past few months - is an AI which manages other AIs.

It isn't "just" sub agents, but you can achieve most of this just with a few agents that take on generic roles, and a skill or command that just tells claude to orchestrate those agents, and a CLAUDE.md that tells it how to maintain plans and task lists, and how to allow the agents to communicate their progress.

It isn't all that hard to bootstrap. It is, however, something most people don't think about and shouldn't need to have to learn how to cobble together themselves, and I'm sure there will be advantages to getting more sophisticated implementations.

Right, but the model is still: you tell the AI what to do, this is the AI tells other AIs what to do. The context makes a huge difference because it has to be able to run autonomously. It is possible to do this with SDK and the workflow is completely different.

It is very difficult to manage task lists in context. Have you actually tried to do this? i.e. not within a Claude Code chat instance but by one-shot prompting. It is possible that they have worked out some way to do this, but when you have tens of tasks, merge conflicts, you are running that prompt over months, etc. At best, it doesn't work. At worst, you are burning a lot of tokens for nothing.

It is hard to bootstrap because this isn't how Claude Code works. If you are just using OpenRouter, it is also not easy because, after setting up tools/rebuilding Claude Code, it is very challenging to setup an environment so the AI can work effectively, errors can be returned, questions returned, etc. Afaik, this is basically what Aider does...it is not easy, it is especially not easy in Claude Code which has a lot of binding choices from the business strategy that Anthropic picked.

> Have you actually tried to do this? i.e. not within a Claude Code chat instance but by one-shot prompting.

You ask if I've tried to do this, and then set constraints that are completely different to what I described.

I have done what I described. Several times for different projects. I have a setup like that running right now in a different window.

> It is hard to bootstrap because this isn't how Claude Code works.

It is how Claude Code works when you give it a number of sub-agents with rules for how to manage files that effectively works like task queues, or skills/mcp servers to interact with communications tools.

> it is not easy

It is not easy to do in a generic way that works without tweaks for every project and every user. It is reasonably easy to do for specific teams where you can adjust it to the desired workflows.

> I can tell you based on your description that you did not do this. Subagents are completely different and cannot be used in this way.

And yet I have used them exactly in the way I described. That you assume they can't just demonstrate that you haven't tried very hard.

> No, it isn't how Claude Code works because Claude Code is designed to work with limited task queues, this is not what this feature is.

Claude allows your setup to execute arbitrary code that gets injected into context. The entire point is that you don't need to rely on built in capabilities of Claude Code to do any of this.

> No, it doesn't work within that context. Again: sharing context between subagents, single instance running for months...I am not even sure why someone would think this could work.

I know what I described works because I am doing it. You can achieve what I described in a variety of ways: Using skills to tell the agents how to access a shared communications channel. Using MCP servers. Just using CLAUDE.md and describe how to use files as a shared communications channel.

This is only difficult if you lack imagination.

> You are talking about having some CLAUDE.md files like you have invented the wheel, lol. HN is great.

No, the exact opposite: I'm saying that this isn't hard, that it isn't anything revolutionary or even special. It's pretty basic usage of the existing facilities. There's no invention there.

You're the one trying to imply this is more revolutionary than it is.

It's natural to assume that subagents will scale to the next level of abstraction; as you mentioned, they do not.

The unlock here is tmux-based session management for the teammates, with two-way communication using agent inbox. It works very well.

> Claude Code has been inherently limited to conversations

How so? I’ve been using “claude -p” for a while now.

But even within an interactive session, an agent call out is non-interactive. It operates entirely autonomously, and then reports back the end result to the top level agent.

I believe they’re talking about Claude Code’s built-in agents feature which works fine with a Max subscription.

https://code.claude.com/docs/en/sub-agents

Are you talking about the same thing or something else like having Claude start new shell sessions?

Okay...and continue to work up the levels? Why do you think OAuth might be limiting? Why do you think they started building subagents first? What is the difference between subagents and products like Aider?

If they were able to wrap the API directly, this is relatively easy to implement but they have to do this within Claude Code which is based on giving a prompt/hiding API access. This is obvious if you think carefully about what Claude Code is, what requests it is sending to the API, etc.

What does OAuth have to do with any of this? I think you are deeply confused.

That’s not what this subthread is about. They’re talking about the subagent within Claude Code itself.

Btw, you can use the Claude Agent SDK (the renamed Claude Code SDK) with a subscription. I can tell you it works out of the box, and AFAIK it is not a ToS violation.

Yes, you can build feature in OP with SDK have done this. Works well...but this is something completely different to agents.

Subagents and the auth implementation are linked because Anthropic's initial strategy was to have a prompt-based interaction which, because of the progress in model performance, has ended up being limiting as users want to run things without prompting. This is why they developed Claude Code Web (this product is more similiar to what this feature will do than subagents, subagents are similar if you have a very shallow understanding...the purpose of this change is to abstract away human interaction, i assume that will use subagents but the context/prompt management is quite different).

Oh really? I was looking at the Agent SDK for an idea and the docs seemed to imply that wasn't the case.

    Unless previously approved, we do 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.

I didn't dig deeper, but I'd pick it back up for a little personal project if I could just use my current subscription. Does it just use your local CC session out of the box?

You can’t resell - that’s the third party language. You can build and use for your own purposes. And yes it just picks up your local sessions out of the box.

> If they gave people API keys then no-one buys their ludicrously priced API product

The main driver for those subscriptions is that their monthly cost with Opus 3.7 and up pays itself back in couple hours of basic CC use, relative to API prices.

You can. This is how Opencode worked, but they are clamping down on that approach.

As someone else has mentioned, you can actually use SDK for programmatic access. But that happens within the CC wrapper so it isn't a true API experience i.e. it has CC tools.

Also created my own version of this. Seems like this is an idea whose time has come.

My implementation was slightly different as there is no shared state between tasks, and I don't run them concurrently/coordinate. Will be interesting to see if this latter part does work because I tried similar patterns and it didn't work. Main issue, as with human devs, was structuring work.

There is an incentive for dishonesty about what AI can and cannot do.

People from OpenAI was saying that GPT2 had achieved AGI. There is a very clear incentive for that statement to be made by people who are not using AI for anything productive.

Even as increasingly bombastic claims are made, it is obvious that the best AI cannot one-shot everything if you are an actual user. And the worst ones: was using Gemini yesterday and it wouldn't stop outputting emojis, was using Grok and it refused to give me a code snippet because it claimed its system prompt forbade this...what can you say?

I don't understand why anyone would want to work on a codebase they didn't understand either. What happens when something goes wrong?

Again though, there is massive financial incentive to make these claims, and some other people will fall along with that because it is good for their career, etc. I have seen this in my own company where senior people are shoehorning this stuff in that they clearly do not actually use or understand (to be clear, this is engineering not management...these are people who definitely should understand but do not).

Great tool, but the 100% vibecoding without looking at the code, for something that you are actually expecting others to use, is a bad idea. Feels more like performance art than actual work. I like jokes, I like coding, room for both but don't confuse the two.

> I don't understand why anyone would want to work on a codebase they didn't understand either. What happens when something goes wrong?

It's your coworker's problem. The one who actually understands the big picture and how the system fits into it. They'll deal with it.

Automatic enforcement of dumb low level stuff is supposed to free up police time for the more serious things. Whether that happens or not is a political decision. I remember the time before red light cameras in London, and the time afterwards, and the situation was much improved after they showed up.

I agree the driving test is too easy (though several orders of magnitude more difficult than in the US states I've had to do one in), and there is too little enforcement of otherwise dangerous behaviour.

I don't think I mentioned anything wrong with automatic enforcement. I think the claim was that when confronted with a financial incentive, people who drive recklessly will stop driving recklessly. Would this be the case if we paid people £50/month to drive better?

It makes no sense at all. The problem in policy is generally that you have people talking past each other: speed limits are effective for people who are generally going to comply with them anyway, they are not intended to stop serious accidents. The majority of accidents are not caused by "accidents" (as most people on here would think them), they are caused by people who drive recklessly a huge proportion of the time and eventually have an accident.

Again, the solution to this is simple: do not give these people driving licences. In the UK, you can kill someone with your car driving recklessly and be out of jail in 18 months. And I don't think people realise this is true, or that this won't have been the first "near miss" for these people...it will have been months and years of doing stuff that will kill someone, and eventually killing them. How are they supposed to kill people with cars if they can't own a car?

Some of these infractions also carry the risk of losing your license. So it is more than just the fine.

> you can look at ONS numbers yourself

ONS numbers say >20,000 fewer frontline officers from 2010-2018, which is pretty much in line with what the union said. See the graph here:

https://www.gov.uk/government/statistics/police-workforce-en...

> In Scotland, there are some days when there is one traffic car covering an area the size of England

Are you high, or did an AI write this?

Area of Scotland: 80,231 km^2

Area of England: 132,932 km^2

So on some days, in Scotland, there is one traffic car covering an area that is larger than Scotland. OK, where's it patrolling? Or are you saying Police Scotland only sends out 60% of one car to cover the whole country?

2010-2018...when did the Tories time in government end? Based on your comment, I am assuming 2018.

Lol, quite the pedant. To be clear though, yes when they are short-staffed they only have one car actually on patrol for the whole country (iirc, the actual full staffing policy overnights is two cars...which you can see has been covered by the media).

Traffic was consolidated into Police Scotland so there is only one police force, and so there aren't local forces patrolling a local area. I believe the total number of traffic police is something like 400 now (which is mostly not people on patrol) and so, overnight around holidays, the policy is to have two cars which turns into less than that on some occasions.

> In Scotland, there are some days when there is one traffic car covering an area the size of England,

Scotland is smaller than England, so this makes no sense.

Furthermore, anyone who drives regularly in Scotland knows this to be completely false - there are plenty of traffic cops around (sometimes incognito too), and they are sometimes even seen waiting in rural and semi-rural areas.

> Hilarious to see pearl-clutching when people point out the SNP has been doing this after complaining about the Tories. This is why the UK is so shit, reality doesn't matter, just politics.

I'm not sure where this tirade came from; I wasn't arguing in favour of any political party.

But getting back to the matter of traffic police - I have eyes, and I can see traffic cops with them. I have family in the force elsewhere in Scotland too, so I know that what you're saying simply isn't true. I really don't know why you are making this false claim.

Right, but what people miss is that some people (usually not academic economists) are able to predict the economy with good levels of accuracy. It is not hard, there is a ton of data, the problem is that you have to drop your own personal interest in politics...for 99.99% of people feeling politically validated is more important than being right on the economy.

Isn't there reams of data on weather patterns too? I thought that like the weather, economic systems are fundamentally chaotic and thus demonstrably unpredictable. There's a whole field of math where we can create these mathematical models that simulate chaos and are fundamentally unpredictable. I was under the impression that Economics falls under the purview of this mathematical theory in terms of unpredictability.

If I may ask, who are the people that do these predictions? What is there methodology and how do you know it's not luck?

The titles have rolled. The prediction in Q1 last year was specifically that by this time, the US would be in a deep recession due to tariffs.

I can understand that most people do not actually stay with these forecasts. The story hits NBC, you are completely outraged about this forecast made by these economists, the thing never happens, NBC has moved onto next disaster coming round the corner, next outrage.

If you work in markets, you are confronted with the relentless inability of talking head economists to just say mildly rational things. It is constant, almost every year now we have this latest economic disaster by economists in the periphery of political parties...no-one pays attention to these people, they have tenure, they make money from cashing in their political contacts not through actual correct forecasts (and yes, they always say that the thing they predicted will definitely happen next year now).

Stopped clock is eventually right. But there was literally zero information in the insane claims made in Q1. If you did not see this immediately, don't pay attention to these forecasts.

That's a really nice comment but does not bear in any way on what I wrote. The Trump administration has now completed its first year. There are three to go.

Yes, i am saying very directly that the prediction made was made by many (which is the comment you replied to) that tariffs WOULD HAVE ALREADY caused the economy to fall into a deep recession. These predictions were wrong, they were obviously wrong at the time they were made but that didn't stop the breathless coverage about a coming Great Depression.

Whatever happens next is irrelevant, it won't make that prediction correct.

There are multiple research papers that indicate that this result (in terms of what you think the paper says) is not obvious. Indeed, to think this is the case, you need an extremely superficial understanding of economics based around "rules" that only exist in theory.

And, if you read the paper, you will find that evidence. How these work depends on initial conditions that vary and exporters will not react in a consistent way.

As a specific example, theoretical research in this area tends to make assumptions around the stationarity of margins that are obviously ludicrous in the context of reality in the US. It is quite easy to justify almost any policy with theoretical research in economics so people who have no understanding of economics will find evidence for whatever position they choose. Reality is quite different.

Great points. This paper is a static, partial-equilibrium analysis that knows the price of everything and the value of nothing, which I suspect the NYT will happily run as gospel.

The most glaring, disqualifying omission is the disregard for FX adjustments. We have not seen the CNY/USD crash yet , but that is because of MASSIVE currency intervention from China to the tune of $200B+ per month: https://x.com/Brad_Setser/status/2012021712012145030

The other wonky thing is they call $200 billion in tariff revenue "a tax" and they also call it a "deadweight loss." Are the authors not the Keynesians I thought they were? At any rate, tariff revenue is not a new tax, but merely a shift in the tax base - since it is $200B that the government does not need to collect elsewhere.

Lastly the collapse of trade volumes from Brazil and India is not a bug it's a feature. Yes supply chains are sticky. The POINT of tariffs is to force supply chain decoupling and reshoring. To "unstuck" the supply chain. Does disruption come with temporary supply shocks? Of course.

The European export growth model is not working for Europe. They are not far behind the US in doing what we are attempting. Canada cozying up to China is not what they wanted for themselves, lol. Etc.

Static analysis in trade is very dangerous. I have never understood why these models are used. These had massive political consequence in the UK during Brexit, close to 100% of the predictions made by these models failed, and they were taken as truth (in the sense that: a prediction was made, and that was reported in the same way as an economic release). There was no real examination of why these known bad models were used.

Calling it a "deadweight loss" also doesn't seem justified by the evidence. It is extremely unclear whether there is a total economic loss because tariffs, if maintained, will have long-term economic consequences. You can point to countries where that went very badly (South American import substitution in the 50/60s, leading to economic free fall by the 80s) and ones where it went very well. It has always seemed quite unclear to me.

The stuff with Europe is also very odd because what do they think sanctions on Russia are? Not just a tariff, a blockade with no revenue raise. But the people who tell you that tariffs are a tax will tell you that sanctions have no issues.

I don't think that tariffs are good either btw. It just seems to a policy choice that is made in a context that can be either good or bad. Europe has massive barriers to trade internally (despite being in a political and economic union) that has been very expensive and harmful because it reduces competition/competitiveness. At least some of the protectionist measures of Trump and Biden are likely to work because the US is fundamentally quite competitive.

The authors ask “who bears the cost of these tariffs?” and use S&P Panijiva data from Jan 2024 to Nov 2025 for their 96% pass through rate to conclude tariffs function “as a consumption tax on Americans.” However, Panjiva is limited to FOIA requests for bill of lading data for 22 countries (Brazil, China, India, Mexico) and excludes major US trading partners, such as the EU, UK, and Canada[1].

The limitations of the data are highlighted by the authors’ somewhat bizarre claim that “a 10 percentage point increase in tariffs leads to only a 0.39% reduction in export prices.” Yet the luxury industry, a major component of European exporters, reduced prices in 35-40% of all products in 2025 and registered a drop in operating margins from 20% in 2023 to 15% in 2025[2]. European car manufacturers also had to adjust. Porsche reported a billion euro 3Q25 loss and a 99% drop in operating profit through 2025, leading to its removal from the DAX and the CEO’s ouster.[3]

The short answer to who pays: too early to tell. Many consumers balk at price increases and reduce consumption, while many foreign exporters seem to be waiting and seeing for the Supreme Court to rule against the administration’s IEEPA claims or the midterm elections before deciding permanent price changes. But exporters are experienced navigators of multiple tariff layers, both internal and external. Many economists have noted the “value-added tax (VAT) system, a tax on final consumption, which the US administration views as similar to a tariff.”[4]

[1] https://www.spglobal.com/market-intelligence/en/solutions/pr... [2] https://www.ft.com/content/9553baf3-7cfc-40c2-b6bd-1871844af... [3] https://observer.com/2025/10/porsche-profit-drop-luxury-econ... [4] https://think.ing.com/downloads/pdf/article/reciprocal-tarif...

I would argue tariffs are inherently easier to deal with than NTBs. I will never understand the absolute hate that is leveled at countries that use tariffs vs. the very noisy, tariff-hating countries that use NTBs.

EU is, obviously, one of the worst offenders here. Tariffs are a great evil...there are still massive barriers to trade within the EU. Let me repeat: WITHIN the EU, a bloc of countries that share a supra-national political system. Like Wymoing putting up a barrier to trade with Iowa (which, btw, do exist in the US too...but are significantly lower than in the EU where there isn't harmonization in even basic products like financial services due to the problems with competition in so many European countries).

The discussions on trade are insane.

Agreed. Never understood all NTB and tariff distinctions, except one suspects the pajandrums in Brussels and Geneva didn't quite believe companies could compete on price or product, and therefore needed a way to keep quotas and subsidies.

I'm not alone in this confusion. During oral argument, several Supreme Court justices asked repeatedly about the distinctions between quotas, tariffs, revenue-raising taxes, non-revenue taxes, etc. in IEEPA's statutory language and precedents like FEA v. Algonquin SNG, Inc., 426 U.S. 548 (1976).

Can you name some of those peer reviewed publications say that tariffs wouldn't be paid by the consumer?

Below is the #1 paper, extraordinarily widely cited. They found that countries with significant market power systematically set higher tariffs on goods with inelastic supply, and in these cases the exporter lowers their price to absorb the tariff, improving the importer's terms of trade.

https://www.columbia.edu/~dew35/Papers/Optimal_Tariffs_Marke...

There was a paper, that i am not going to dig out, on the first set of China tariffs by Trump that found that exporters reduced margins. You don't have to look far, you just need to look a few years ago. For some reason, the claims made earlier last year were apocalyptic despite there being clear evidence from the very recent past that this wasn't the case (making money in equities last year was terribly easy).

Btw, you can find papers where tariffs are paid by consumer, where they aren't, where there is no effect. The thing that is being measured is completely different to the actual tariffs. The impact, like everything in economics, depends on the context in which the tool is used. That context is typically hard to model, so we end up with a lot of shitty papers claiming that it is about the tariff when it is about the context/implementation/etc.

No, completely wrong. Deep markets are not more accurate.

People who are unfamiliar with how regulated gamblings works assume that the "market" is just lots of informed people rationally trading with each other. This is not how it works. Bookmakers post lines to a small group of syndicates up to a limit, they will often do this non-publicly, and this is how prices are set. They are not set by the "wisdom of crowds", they are set by people who have invested hundreds of millions of dollars in predicting the outcome because bookmakers have an economic need for accurate lines.

When lines open to the public, there is often no significant movement after opening prices set by syndicates. That is because the public has no idea what the actual price should be, they are just uninformed noise traders clicking buttons randomly...that is the product too, the purpose of the product is entertainment not economic efficiency.

It is true that some lines are set incorrectly but the public is not able to benefit from this, because they do not have the information. I would guess that 95% of money made from gambling has been made by under 50 people. And, perhaps counter-intuitively, most of the time these people trading does not have an impact on price because they deliberately trade in a way that does not impact price. Again, the purpose is the same: they trade to make money, not produce economic efficiency.

The people who think prediction markets are useful in any way are people who never traded any markets and couldn't predict if the sun is going to come up tomorrow. If gamblers are noise traders, these people are noise speakers. These markets are completely pointless, gambling is economically pointless outside of the pleasure that people get from entertainment.

Noise traders ultimately create an even greater incentive for accurate prediction. The fact that the odds are set at the start and never change just proves that there's very little change in relevant information about upcoming sport games, races etc. where regulated bets happen. That's totally normal. Bets about real-world events are a rather different matter though.

No, they don't. If you have a line that is beatable, that line has been open for a long time, and you have informed people profiting from that line, it will usually not move. People who have information will disguise their flow, they won't bet with places that will move the line against them when they bet (if you bet this with Pinnacle, for example, they will work out you are beating their line immediately, they have quants who can work out how you are beating their line, and you have permanently destroyed your edge) so you put money down at soft book somehow and they will likely not move line against you...meaning the line doesn't move.

Again, it is fairly common assumption that people make that it must be noise traders who are incentivizing syndicates. This is the case at open but not after, and there is a significant distinction between noise traders and noise traders through retail books. Retail books do not set the lines, they do not post lines early to syndicates, their product is completely different. There is literally no incentive for accurate prediction because the economic gain from noise trading does not accrue to anyone who has information. 95% accrues to firms with the greatest marketing advantage, again...this is entertainment, it is not about accurate prediction, you are misunderstanding at a very fundamental level what is going on here. It is like going to see the Minecraft Movie and thinking this is artistic expression on the level of Tokyo Story.

This would all be different if there were real markets underpinned by economic demand for this risk but there isn't. This is why Betfair failed. This is why these "prices" you see aren't actually real prices.

Are you describing how sports gambling odds are set? Prediction markets Polymarket and Kalshi do not function the way you describe, at least, not yet.

> People who are unfamiliar with how regulated gamblings works assume that

I work in the regulated gambling industry. Prediction markets like Kalshi and Polymarket do not work like the regulated gambling industry. You don't know what you're talking about.