It's Thariq from the Claude Code team here. This was my change! I made the AskUserQuestion tool so am generally in charge of maintaining it.
First, overall wanted to apologize and agree that this did not meet our bar and does not represent how we plan to ship on Claude Code.
To give you a motivating sense, as the models get more powerful, usage patterns start to change. I'd gotten a lot of feedback that AskUserQuestion tool was starting to block some long running jobs unexpectedly and so I tried a change to help that.
Our internal feedback on this was good, but the rollout should have been opt-in (like it is now) and on the Changelog.
Thanks for the feedback! We're always trying to make Claude Code better while balancing it with how people use it in many diverse ways. I did not really intend AskUserQuestion to be a safety gate when I first built it, but I realize it has evolved in that direction for some users.
I'm still exploring other ways of helping with this problem of balancing longrunning work and input, but will take lessons from the rollout here.
Instead of one-off fixes Claude should have a much richer interface to configure between "ask approval every time" and "YOLO dangerously". I should be able to trivially set "run this task until completed" and have settings like: don't consult the web, don't touch files outside of the codebase, don't delete anything, etc. They don't have to be perfect, just better than the all or nothing system we have now.
Codex legit does a great job of this with it's auto-review feature. Does a good job of understanding what permissions I'm implicitly giving with the request, and where it needs to actually ask me to grant permission.
Until the classifier is wrong or also prompt injected. the classifier is just as vulnerable as the model itself is. Yes it is harder to break but trying to make a nondeterministic tool deterministic by adding another nondeterministic one on top just reduces the chance of something going wrong.
Tbf as long as that chance is low enough it doesn't matter in practice, but I have definitely seen the classifier approve things that were questionable, and I've also seen it decline things that were obviously okay.
It would be nice to use automode in planning mode.
Right now planning mode (e.g. I want claude to do all the planning necessary to create a robust plan file) still prompts you with trivial things, especially one-off things you wouldn't care to add to your whitelist. Yet if you switch into automode, then it's not in planning mode anymore.
Repro:
1. Prompt "Implement a hello world in C" in planning mode
I feel like this could be solved with an NLP interface to SELinux, spinning up policies on the fly, blocking network access or giving read only permissions on a per-conversation basis.
This is by design. Anthropic treat users as paying workers (how futuristic!). Their end game is to replace software developers (and it's not like they are trying to hide that?)
I read the article but I find this response kind of strange. Am I alone in this?
Wanton accountability for a multi-billion dollar cutting edge company… leaves more to be desired from the best? Take Apple or Google or any top tech company of the past at its prime and compare. This kind of behavior then would probably reflect poorly on the institutions behind the tech and not maintain their image of technical brilliance because it shows weakness in a vulnerable way. It is human. It is not strategic.
By wanton accountability, I mean things like saying “This was my change!” or “ wanted to apologize and agree that this did not meet our bar and does not represent how we plan to ship on Claude Code”. It makes them (the company) look weak? Accountability is important but where, how, and when you do it is even more important. These stakes are, not joking, life and death, and in this big game of chess we get paid for, naivety not only in our technical implementation can weaken our position.
Not trying to attack, just trying to learn and probe with the community. Maybe a cost of this kind of transparency on the internet. I am wondering if this a new trend and tech companies are changing in a way I don’t understand! In any case, it’s really cool work that is being with Claude Code!
> it shows weakness in a vulnerable way. It is human. It is not strategic.
Are you complaining that the answer is too human and that a multi billion company should not allow a human who made a mistake to own the mistake in public, being honest about what happened in this case?? Would you prefer complete silence from them like you most certainly would from Google or Apple.
That sounds incredibly sad to me, we don’t even expect humanity from big tech since that’s what we’ve become accustomed to seeing.
fTR both the Anthropic dev response and the blog post seem to believe that a single person can be blamed for something like this, which I wholeheartedly disagree with! Nobody reviews your changes? There’s no QA? Not even an AI checking the release notes match the diff from the previous release?? Blaming a dev for “putting a serious bug in production” sounds really 90’s to me.
> It's Thariq from the Claude Code team here. This was my change! I made the AskUserQuestion tool so am generally in charge of maintaining it.
In a sense yes, I think it is actually reasonable to complain that the answer is too human/individualized here because it likely wasn't this individual human who made this decision, but he's making it seem like it is so that we are less likely to blame the company as a whole.
It's counterintuitive but when one singular person owns up to the problems that, at the root, are actually systemic to the decision making of the whole company it plays on the psychology of us as humans.
"I'm in charge of maintaining it" - This is not the same as "I'm in charge of all of the decision-making behind the implementation of how this tool works for users".
I actually agree exactly with your last point that one single person taking blame is counter-intuitive/non-productive here, but it actually seems like what these large companies desire is to have one person be the fall guy to play on people's sympathies.
If this were some small startup it would make sense but this is not that case.
It’s also an unforced error in my eyes for a multi-billion dollar company with competitors in the US and outside. Mistakes happen but no reason to turn a mistake into a revelation about processes that leaves no ambiguity, one that leaves the mythos (pun) of your company less than before.
Unfortunately this is a sign that the systems by which PRs are shipped need serious inspection.
For example in incident management, the industry settled on blameless postmortems. Blame is pointed at the systems which allowed incidents to occur, never the human(s) who triggered the incident. "Just don't make that mistake again" simply doesn't work. Humans make mistakes!
There are too many stakeholders in a product this large, even if every one of them wishes it were not so. The systems by which this PR shipped need serious inspection.
I don't really care, in a most positive way towards you.
I don't hold you accountable even if you twirled your mustache like a villain. You are not responsible for Claude Code. Why did the leadership set up such a weak, error-prone process?
Again, not asking you directly, because it doesn't matter to me what you say, respectfully.
There has to be a second PR approver. There's no way a company as large this can allow an individual the ability to push to production without a secondary person involved.
When there's at least two people involved you can then start to look at more systematic factors that go beyond any single human mistake.
When both the submitter and approver miss that there's no changelog entry for a PR does that mean a checklist is needed? Should the changelog be automated with the commit message which can be improved for external consumption?
I'd find it very hard to believe that this high profile change just happened to be the only change missing from the changelog.
Bugs are always going to happen but housekeeping like writing something to a changelog when a commit is merged can be quality assured.
If there really is no PR reviewer and no separate QA what is Anthropic even doing. I have respect for someone owning their mistake but as others have mentioned it reveals what may be a systemic issue.
Seems strange for a company of your size to have one person push changes that should have easily caught this edge case then. Seems like a change even a small handful of people could have reasonably thought up this side effect of.
This is the kind of weakness I was alluding to… internal information that shows dysfunction is now shown to the public and there is no ambiguity. People pay attention and there is a low tolerance for unforced errors.
Habibi, I don't think anyone is really blaming you personally. We are piling on to the fact that this piece of critical infrastructure many of us depend on day in day out is being built in such a way that a single developer can wake up one morning, ship a change they thought might be nice, and then walk it back the next day.
Hey, mistakes happen. As we saw with the recent AWS billing issue. But Anthropic is a leading company that makes cutting-edge AI which will kill people. It’s literally a matter of life and death and at this point, it is the place to be if you want to be the best. So as an institution, I don’t want human answers, I want the right, most professional answers. We used to have it. Human responses with every word chosen to maximize every aspect that needs to be maximized, by engineer grey beards no less! And maybe they don’t want to have it. That’s why I made my original comment. Maybe it’s a strategic decision to allow this kind of accountability in a public forum. I think it’s wanton but others might find it refreshing.
And to clarify why it even matters to me is, these firms are in competition within the US and outside. Every decision by every person has an effect. Moves like this have a cost. I’d have imagined that at the level these companies operate, any public facing engineer only communicates correctly with the proper perspectives in mind... We have more data points now we didn’t have before.
To drive the point home, now we and every other competitor knows, for free, processes that are going on at Anthropic, that in my opinion make them (as a company) look weak. I
I expect the person responsible for Claude Code to explain the reason this direction was chosen.
It is ridiculous to expect individual contributors to do so.
Where is the leadership? Can individual contributors decide to significantly alter the workflow of millions of customers? Were any customers consulted at all?
> we don’t even expect humanity from big tech since that’s what we’ve become accustomed to seeing.
There is a difference between humanity in communication and saying the problem was a human choice.
You can have humane communication about a system failure and actors in that system. Or you can have cold matter of fact press release voice about an individual persons failure.
I don’t think anthropic statement was any less distant than other times big tech brushes off a whiopsie. They just used personal language to ingratiate the reader.
It doesn't feel strange to me at all. It feels like a very human response from the person that introduced the change. I great appreciate that, rather than ignoring the problem, or some canned corpspeak response.
Hi Thariq, Chris from ApprovIQ here. We depend on Claude Code to ship. It's our most crucial tool.
But we feel an asymmetry. We feel it's not an important to Anthropic as it is to us. We feel the systems by which PRs ship on your side need serious inspection.
Claude Code is career-level important not just to us but to a lot of people. So actually, your post and claim of individual accountability gives me less confidence.
For a product of this size and a comm to its core audience after a breach of trust like this, from a company of Anthropic's size and with as high quality a video production department and that produces as much research as it does, "this was my change" is quite discordant.
My hope was we'd have someone with an organizational mindset admit to the organizational failings that allowed this to happen. It would be great to see that.
For what it's worth, I totally understand the motivating use case here. There were absolutely times where I walked away from what I was hoping would be an hours-long project that would run to completion and came back to find that Claude had asked me a question early on and I'd missed out on a large amount of implementation time. So you were not imagining that the use case is real!
It's also worth adding that I really enjoy the AskUserQuestion feature and will regularly ask Claude to specifically use it instead of asking me questions in plain text because it's a lot easier to work with.
It's always good to learn from mistakes, and I appreciate both your work on this and you coming here to own it. Keep up the good work!
Is there a way to turn options-based AskUserQuestion off? I couldn't find it at all, and the "options" selector is the most annoying thing in CC for me, plain-text is the only way, everything else is distracting. I know, I can use `n` (sometimes) or cancel it, but both bring more pain that just regular communication (cancellation does one more chat step + requires an extra action, same with `n`)
Are you also responsible for AskUserQuestion swallowing the preceding output on Fable? It asks me questions about its response that it never even showed me. It's (and I hate to use the term) unusable.
It is such a shame that Anthropic has no interest in QA, because they have incredible models, and bafflingly broken products. In an alternate universe Claude Code would be 10x better than where it is now...
hello, please fix needing to reauth every day (sometimes with email verification). This started happening this month. It's tedious and makes switching very tempting.
It isn’t very smart to recover the code, and Anthropic will ban you because, apparently, decompiling code or training a model constitutes cybercrime, according to Dario and his lackeys.
(Edit: BTW, if you haven't, read Nudge by Thaler and Sunstein. It's a somewhat long-winded exploration on choice architectures, but you're neck deep in that space now. I'll give you my copy if you're in SF.)
Thanks for the openness. I got bit by this one and was, frankly, pretty surprised.
The funny thing about user-facing interaction mechanics is that everyone is part of some minority, and everyone comes with their own sense of what "natural" or "obvious" is. With something this impactful, communicated clarity of behavior will important. Your feature is also doing double-duty, serving as a last net against prompt-injection attacks by giving the user the final say.
(Also, BTW, folks outside of Anthropic are unlikely to be as tooled-up for long-running unsupervised Claude jaunts as you guys. The cost of wild success is wide adoption.)
One thing I'll suggest is that the mechanics of permissions and asking are presently pretty hacker/nerd friendly but simultaneously too-scary and not-scary-enough for non-coders.
Examples:
- Wild-cards on always approve is awesome, but, with prefixes like timeout and nohup, the "thing" that is getting done is buried and largely unexplained to the user.
- Auto is actually kind of a sweet spot (sometimes goes off into the weeds), but the designers and PM's I've been working with might as well YOLO. They have no idea if they're breaking things, but they gravitate between plan and auto mode.
- Fewer permission prompts is great, but it comes after a user has slogged through generation of a data-set to work against, like battle scars for paper cuts. It's the thermostat problem. The signal comes when the user is uncomfortable. And it's a way to learn me, but not me now.
I've had good fortune with Opus 4.8 and Fable just telling the system what phase of my life it's in. Things like "I'm going to go make dinner... Go profile the matrix or configurations and build the dataset for the next two hours while I'm away" have a pretty good hit rate. On the flip side "keep me in the loop and bring me your results before making structural changes" also articulates well with Fable. It will tread more carefully.
And these approaches are the ones we'd use with someone transitioning from SDE1 to SDE2. A little more autonomy, and the grounding in the bigger picture. Can we eventually translate to perfectly judging what the user wants in the moment based on incomplete signal?
No, but I'm glad you're trying. Keep the interaction model clear to your broad set of users, and we'll come along for the ride.
> Not every feature will necessarily appear in the changelog
This was such a frustrating part of this incident, along with Anthropic's refusal to explain why the changelog is no longer a complete record, what else is going out? [1]
Boris Cherny's only participation in the thread was to delete "extreme danger" from the GH issue title [2]
I guess we should be thankful they added an option and disabled it by default. OpenAI is standing firm on their decision to not allow their 60s timeout to be disabled, [3] however more of the Codex harness is open source so customers have been able to fork it to add the option themselves.
Every time they deploy a new model, it feels like older models get slower or dumber. Maybe because of resource allocations, maybe because they started to use quantized models, cant be sure of that.
What i found is that you have to adapt your workflow to the specific model you are using and readapt it every time you notice some problems. Whether of opus4.8, or sonnet5.
As soon as tokens stop being subsidized I would not trust any harness made by a company that also charges for the compute.
Right now the interests align, but as soon as more tokens -> more profit (instead of more revenue and more losses) the perverse incentives will be too big to avoid.
It seems inevitable that open source harnesses will win. These companies would do better to just open source their harnesses.
An OpenAI subscription seems like a decent alternative for people using open source coding agents. It has usage limits too, but for $20/month it’s not bad.
The headache I recently had was it somehow started interpreting mouse clicks in the terminal to mean I clicked an option when I was really just trying to get/confirm window focus.
They also randomly changed how copy-paste works in a way that made it really annoying to copy text.
Vibe-coders pumping out features: this is the reaction you're going to get if you inflict this on your users! Nobody wants software that updates every day and changes in fundamental ways that are hard to keep up with.
Copy/paste is one of the most basic, low-level features of a modern operating system. NO APPLICATION SHOULD EVER SCREW WITH IT, IN ANY WAY!
And I say this not just as a seriously annoyed user, but also as a professional UI developer: it is a well-known anti-pattern to override the user's expectations, at any level ... and that applies tenfold to the most basic patterns that every other app follows.
I don't care if you added a magic way to write all my code for me: if the only way to invoke it is to break copy/paste, you've failed at development!
> Copy/paste is one of the most basic, low-level features of a modern operating system. NO APPLICATION SHOULD EVER SCREW WITH IT, IN ANY WAY!
-Ghostty enters the chat-
Middle-click paste in Ghostty is hazardous: the damn thing can move your cursor to the position of your mouse pointer and paste there. An utterly infuriating behaviour - it is a text terminal. It has no business emulating the usability crimes GUIs first committed and then committed to.
I wonder if it's the same behavior as copilot cli. I dug into it when I got annoyed and found that there's a fundamental tension between rendering the cell based UI and supporting the terminal's native copy paste. At least on copilot cli there were extra zero width characters being copied.
My memory is that once you begin using the alternate screen, you need to turn on mouse reporting, and then you lose native copy. Different terminal emulators also act totally differently when in this state.
This bit me too. If I wanted a mouse-driven app, I'd use the GUI. I don't understand why they're trying to replicate that experience in a text environment.
I saw that but it asked me if I wanted to optin first! I did accidentally and hated it since I always click on the terminal to make it focused and now accidentally click on something that actually responds!
I really hate that change. It's counterintuitive that a shell recognizes mouse clicks to mean "select this option".
Also, the auto-copy selected text feature is awful. That's another thing that is so counterintuitive. I have never seen a UI where just selecting text means auto-copy to the paste buffer. Who is coming up with these awful UX ideas?
On top of that, the scroll "feel" of the new client (at least on macOS) isn't quite as good as their old one. It just feels off and not like a native terminal.
Both Claude Code and Codex have issues overwriting or duplicating text for me when scrolling.
Probably from trying to keep scrollback virtual while anticipating terminal resize. But I wouldn't mind a way to opt back in to naive, unlimited convo scrollback.
One of the things I'm baffled about with Claude code is it seems like it setup a really chunky VM on my computer, and yet by default it doesn't seem to sandbox anything. Also the newer models seem really aggressive about modifying your computer. Yesterday Claude started rewriting system files on my linux machine (user accessible, but still way outside the scope of what I asked.) This wasn't me asking it to debug my machine, I was asking it to debug some frontend UI code. Once I put it in a sandbox I started realizing how often it tries to poke out of it for really lame reasons.
More reason for me to use OpenCode and my local LLMs
It happened to me today. I was reading agent's answer and it asked me something. I didn't even get to the question - it accepted something! Jesus Christ. Where are the software engineers?!
I’ve seen that a few times. Had to disable the TUI mode because clicks to focus the terminal window were being interpreted as an approval even though the click was nowhere near the question being asked. I’ve also never chosen to enable auto mode but somehow it is on and approving shell commands I didn’t want approved. Scary since I’ve caught it adding things like auto approve flags to terraform apply commands.
I’d really suggest to run Claude Code in a sandbox! Anthropic can’t be trusted.
I run it on Mac with sandbox-exec. No access to AWS, database, pretty much just files in the current folder. Even if it goes rogue, it can’t do much harm.
A while back, I saw a similar feature land in Codex (I'm using the VS Code plugin) but it got removed quickly. What is the chance that an LLM recommended this same idea to the Claude PM or lead? I see LLMs across different providers converging on similar ideas or biases frequently.
So much attribution to malintent here, but most likely they're trying to build a product with the features that they themselves would use, and from my own experience it's very frustrating to leave a Claude session running and come back to find it did nothing because it got stuck on a question.
Furthermore, believing that the only thing saving you from disaster is Claude deciding to ask you a question is not a great conclusion either. You need guardrails in the power you bestow upon Claude from outside, not from inside.
Meanwhile, this article was written by Claude and has sentences like "Which cuts less far than it looks.", which I doubt Claude stopped to ask about.
The prose was written by me, with the research being done by Claude and also clearly attributed. I left Claude's research as a series of bullet points so that it would be clear that I'm not passing off an LLM's work as my own, but if anyone wants to dig deeper, they have some starting points to consider.
I don't publish prose written by an LLM for the same reason I would not have an LLM solve a crossword puzzle for me -- there's no joy in that.
The article was obviously written by an LLM, it's obviously from the overall structure as well as individual sentences. Maybe you've been exposed to so much LLM text that you write identically to one, in which case, yikes!
No worries. :) I did struggle a little with how to make it obvious that the machine output was not my own. I set it off inside horizontal rules and I did add some prose before and after to say when Claude's output began and ended. If anyone just jumps around using the table of contents, they'll miss that, though. (Not saying that was the case for you).
I had considered putting the research into a gist, but I don't want to have to rely on a 3rd party to keep something that's integral to my post online. Seemed more cohesive to keep it all in one place.
What I was trying to get across was that I had some basis for the points I was trying to make and wasn't just handwaving about it, but I also didn't want to spend endless hours whittling it down. We're all busy people and I see now that it's entirely possible to skim the post and assume it's some low effort word salad. It may be a word salad, but it wasn't low effort. ;)
Come on now, arguing it's a good feature is different than arguing that it should be turned on by default for everyone with no documented option to turn it off.
I like this feature as an OPTION that is DOCUMENTED. But that's not how it was rolled out. The CC team + processes clearly failed here. And this is far from the first example of this kind of slopiness.
I don't think it was malicious, but I do think it was quite obviously reckless/careless, and I feel like most of the comments here match that sentiment. Everyone and every team makes mistakes, but unfortunately I don't see the CC team really learning from those mistakes even though they keep happening.
CC is not a stable or reliable tool, it is bleeding edge, and that's a tradeoff you make when you pick it over other harnesses.
> and from my own experience it's very frustrating to leave a Claude session running and come back to find it did nothing because it got stuck on a question.
I cannot fathom implementing and shipping a feature to hundreds of thousands of people without even asking basic questions like: "what types of questions does Claude ask users".
Literally one of the most used plugins in their entire ecosystem, provided via their official plugin marketplace, is Superpowers. A plugin whose very first operating step is _asking numerous questions about product requirements_. Of course those prompts can't be skipped!
It wasn't even parameterized for Claude to tell the prompt what severity of question was being asked to allow at least _something_ to categorize urgency or expected response time.
Even more egregiously, 60 seconds!? The first time I noticed this happened was when it asked me a question, I turned to my second monitor to go look at some product documentation to get an answer, and by the time I turned back it had skipped me. How can I possibly provide any kind of informed answer in under 60 seconds? I can barely read some of its context for a question in 60 seconds!
I don't think they did this with malintent, but I do think this shows an enormous gap in judgement in how they handle the idea to delivery pipeline.
Yeah the UX doesn't seem great, though I do think in general the AskQuestion tool is very awkward when it gets used in a very long running process (I personally use Claude to run repetitive tests to slowly iterate on a problem and verify in true conditions that it was solved), so I wouldn't like it without the auto-continue either.
(I had originally replied on safety of letting LLMs skip questions but I don't think that was your point so I removed it.)
Seemed fine to me. It’s ask user question not the permission gate. Maybe there should be a new feature enabling warning or something but I think this is the right default. The models are good enough to just proceed with an option of their own and then you can go correct them afterwards. Makes fleet management easier.
Anthropic's willingness to completely change how claude code works is one of the stated reasons why the Pi coding agent exists. When people are dedicating significant time to building workflows on your product, consistency is very important.
Yeah, when I saw this in the Claude Code CLI I was completely baffled. I've been using the thing through their agent SDK for several months now so I wouldn't have to deal with any of the wonky shit they change every second week in the CLI.
The tool seems to be covering three different things: auth, missing info and preference - and they should not share one timeout policy and all of that.
I really hate this direction both Anthropic and Open AI are following. They are in this silly competition whose model/harness can go unattended the longest, no matter what. And it is never explicit, you learn about it after you get bitten by it. Claude Code has auto mode which is supposed to take over permission prompts but no, they had to couple that with “I will assume this is what the user wants” and made it unusable.
It's what some more experienced users want and the companies are following the well-trod path of optimizing heavily for power users at the expense of complexity, only now it's gotten easier to add absurd amount of code to a project. Not necessarily to make it work right. Personally I have some tasks where sessions between one and five days are typical so I appreciate that it's possible.
Experienced in what dimension? I've been using these tools for about a year and coding for 20+ years, and frankly these long horizon tasks are the OPPOSITE of what I want. I want quick iteration cycles so that it doesn't spend a lot of time and tokens building things I need to throw out. I think the people that mostly want long horizon tasks are: AI labs, because they want you to spend tokens, and vibe coders, who are mostly using it for entertainment purposes.
I don't know about the overall breakdown but in my case longer runs are prototyping, bug hunting, reverse engineering, etc. For example Gnome Remote Desktop didn't work in my configuration due to a combination of hardware and codec bugs and settings. One drive to make it work, another to backport the current upstream packages to Debian stable, restack the patches, and push to my machines. Another sequence of long runs was writing a new client for a closed-source conferencing service I use to allow fixing some particularly irritating bugs. Exploit development has the same shape although that's not something I do personally. From what I've seen the amount of useful hands-off run time is directly related to how clearly it's possible to specify a concrete, verifiable standard by which to judge the outcome. For some tasks that might be days or weeks, for forward engineering on a software product that for me is usually under an hour.
I work on systems code that is tricky to write, but with a good test harness. Being able to leave the agent unattended to try several paths when I'm generating the first draft of some code is very helpful.
Of course beating the code into shape for submission requires more manual work. But the draft stage is valuable to find unexpected friction points.
I would love Claude Code to be a little less vibe-coded. The underlying model is excellent, but we're being pretty much forced into using CC to use the subscription model.
I am a longtime and heavy Claude Code user, but Anthropic's product management overall (including for their Desktop/web products) has been really baffling me. I agree that these things often have the air of having been vibe coded without enough human input, and change so quickly (often without a particularly compelling reason for the change) as to be aggravating.
The most recent one that's had me annoyed is the "Fullscreen" TUI feature, which is super unintuitive, implementing its own text highlighting and copy-on-select mechanics, overriding your terminal's native right click. Easy to disable but terrible defaults, IMO. It's not even really clear to me what problem it was actually supposed to solve.
Yah. And it's not like they can't afford the talent to do this right either. I've said elsewhere, I think it's an attribution error. Claude Code is massively popular, but arguably because of the model/subscription, but I think the brass reads this as "great success throughout"
> What if the agent makes the wrong choice? How many tokens have been burned in the meantime?
It is much worse than that. Claude Code doesn't auto-commit when stopping for an answer. There might be possible data loss if an uncommitted file is edited.
Good luck recovering the file from the JSONL conversation history.
On the other hand, relying on Claude Code's internal version control puts you at the same mercy of their product decisions and move-fast breakage.
Instead, start with a plan file and tell the agent to break it up into logical commits.
Though I think the bigger issue here is when you're yoloing something mutable, like managing a remote server or driving a browser or troubleshooting your local OS where there's no going back.
It's actually pretty straightforward to recover file-states from conversation history. I accidentally deleted the wrong repo on my machine once and recreated all the lost work from agent chat history. It is, ironically, the sort of task which AI agents excel at.
Anthropic makes more money when people use 5% of what their subscription offers them. This allows them to sell more subscriptions without paying for more capacity.
I don't think the current subscription price is intended to be a money maker. It's the loss leader to get people invested in the companies' tooling, and make those people more willing to start paying higher enterprise rates as they grow.
Do we have any concrete numbers of how many of their users are subscription vs enterprise, though? Because enterprise users are paying API prices (or at least my employers are)
I think this is a good feature, but should be gated behind a toggle that is off by default, and designed to be enabled per session via prompt.
There are situations when I want Claude to start working on something just as I'm about to head to bed or otherwise step away. It's kind of annoying to come back only to find that Claude worked for just 5 minutes and then decided to pause and ask a question.
That said, I think certain types of questions should not be automatable. Maybe it's already built that way, but I wouldn't want Claude to go with its recommended direction for anything related to operations like deletions, changing external systems, etc. Basically, things that cannot be undone should be a hard-block and wait for user input always.
we need a real agent that represent our own interest and deal with a more powerful general agents. And we can delegate answering this question to those more personal agents.
Yeah it's exhausting to read through, but still made it onto the front page with overwhelming upvote/comment margin. I guess people either only read the title and/or don't care about AI slop if the underlying thesis is compelling.
I built in Human blockers into my agentic workflows with great intention, so naturally this annoyed me deeply. 14d ago, I was elated to see a random blog post that detailed the “fix” [0].
Hi everyone,
It's Thariq from the Claude Code team here. This was my change! I made the AskUserQuestion tool so am generally in charge of maintaining it.
First, overall wanted to apologize and agree that this did not meet our bar and does not represent how we plan to ship on Claude Code.
To give you a motivating sense, as the models get more powerful, usage patterns start to change. I'd gotten a lot of feedback that AskUserQuestion tool was starting to block some long running jobs unexpectedly and so I tried a change to help that.
Our internal feedback on this was good, but the rollout should have been opt-in (like it is now) and on the Changelog.
Thanks for the feedback! We're always trying to make Claude Code better while balancing it with how people use it in many diverse ways. I did not really intend AskUserQuestion to be a safety gate when I first built it, but I realize it has evolved in that direction for some users.
I'm still exploring other ways of helping with this problem of balancing longrunning work and input, but will take lessons from the rollout here.
Instead of one-off fixes Claude should have a much richer interface to configure between "ask approval every time" and "YOLO dangerously". I should be able to trivially set "run this task until completed" and have settings like: don't consult the web, don't touch files outside of the codebase, don't delete anything, etc. They don't have to be perfect, just better than the all or nothing system we have now.
Codex legit does a great job of this with it's auto-review feature. Does a good job of understanding what permissions I'm implicitly giving with the request, and where it needs to actually ask me to grant permission.
automode mostly fixes these things, it runs a classifier on every request that would have required permissions to make sure it matches your request
Until the classifier is wrong or also prompt injected. the classifier is just as vulnerable as the model itself is. Yes it is harder to break but trying to make a nondeterministic tool deterministic by adding another nondeterministic one on top just reduces the chance of something going wrong.
Tbf as long as that chance is low enough it doesn't matter in practice, but I have definitely seen the classifier approve things that were questionable, and I've also seen it decline things that were obviously okay.
It would be nice to use automode in planning mode.
Right now planning mode (e.g. I want claude to do all the planning necessary to create a robust plan file) still prompts you with trivial things, especially one-off things you wouldn't care to add to your whitelist. Yet if you switch into automode, then it's not in planning mode anymore.
Repro:
1. Prompt "Implement a hello world in C" in planning mode
2. Switch to automode
3. It writes hello.c instead of a plan file
I feel like this could be solved with an NLP interface to SELinux, spinning up policies on the fly, blocking network access or giving read only permissions on a per-conversation basis.
At least there's an auto-approval mode now that uses another agent to sanity check commands. Before it, the options really were manual vs yolo.
This is by design. Anthropic treat users as paying workers (how futuristic!). Their end game is to replace software developers (and it's not like they are trying to hide that?)
I read the article but I find this response kind of strange. Am I alone in this?
Wanton accountability for a multi-billion dollar cutting edge company… leaves more to be desired from the best? Take Apple or Google or any top tech company of the past at its prime and compare. This kind of behavior then would probably reflect poorly on the institutions behind the tech and not maintain their image of technical brilliance because it shows weakness in a vulnerable way. It is human. It is not strategic.
By wanton accountability, I mean things like saying “This was my change!” or “ wanted to apologize and agree that this did not meet our bar and does not represent how we plan to ship on Claude Code”. It makes them (the company) look weak? Accountability is important but where, how, and when you do it is even more important. These stakes are, not joking, life and death, and in this big game of chess we get paid for, naivety not only in our technical implementation can weaken our position.
Not trying to attack, just trying to learn and probe with the community. Maybe a cost of this kind of transparency on the internet. I am wondering if this a new trend and tech companies are changing in a way I don’t understand! In any case, it’s really cool work that is being with Claude Code!
> it shows weakness in a vulnerable way. It is human. It is not strategic.
Are you complaining that the answer is too human and that a multi billion company should not allow a human who made a mistake to own the mistake in public, being honest about what happened in this case?? Would you prefer complete silence from them like you most certainly would from Google or Apple.
That sounds incredibly sad to me, we don’t even expect humanity from big tech since that’s what we’ve become accustomed to seeing.
fTR both the Anthropic dev response and the blog post seem to believe that a single person can be blamed for something like this, which I wholeheartedly disagree with! Nobody reviews your changes? There’s no QA? Not even an AI checking the release notes match the diff from the previous release?? Blaming a dev for “putting a serious bug in production” sounds really 90’s to me.
> It's Thariq from the Claude Code team here. This was my change! I made the AskUserQuestion tool so am generally in charge of maintaining it.
In a sense yes, I think it is actually reasonable to complain that the answer is too human/individualized here because it likely wasn't this individual human who made this decision, but he's making it seem like it is so that we are less likely to blame the company as a whole.
It's counterintuitive but when one singular person owns up to the problems that, at the root, are actually systemic to the decision making of the whole company it plays on the psychology of us as humans.
"I'm in charge of maintaining it" - This is not the same as "I'm in charge of all of the decision-making behind the implementation of how this tool works for users".
I actually agree exactly with your last point that one single person taking blame is counter-intuitive/non-productive here, but it actually seems like what these large companies desire is to have one person be the fall guy to play on people's sympathies.
If this were some small startup it would make sense but this is not that case.
It’s also an unforced error in my eyes for a multi-billion dollar company with competitors in the US and outside. Mistakes happen but no reason to turn a mistake into a revelation about processes that leaves no ambiguity, one that leaves the mythos (pun) of your company less than before.
I did make the decision, and shipped the PR.
Unfortunately this is a sign that the systems by which PRs are shipped need serious inspection.
For example in incident management, the industry settled on blameless postmortems. Blame is pointed at the systems which allowed incidents to occur, never the human(s) who triggered the incident. "Just don't make that mistake again" simply doesn't work. Humans make mistakes!
There are too many stakeholders in a product this large, even if every one of them wishes it were not so. The systems by which this PR shipped need serious inspection.
I don't really care, in a most positive way towards you.
I don't hold you accountable even if you twirled your mustache like a villain. You are not responsible for Claude Code. Why did the leadership set up such a weak, error-prone process?
Again, not asking you directly, because it doesn't matter to me what you say, respectfully.
There has to be a second PR approver. There's no way a company as large this can allow an individual the ability to push to production without a secondary person involved.
When there's at least two people involved you can then start to look at more systematic factors that go beyond any single human mistake.
When both the submitter and approver miss that there's no changelog entry for a PR does that mean a checklist is needed? Should the changelog be automated with the commit message which can be improved for external consumption?
I'd find it very hard to believe that this high profile change just happened to be the only change missing from the changelog.
Bugs are always going to happen but housekeeping like writing something to a changelog when a commit is merged can be quality assured.
If there really is no PR reviewer and no separate QA what is Anthropic even doing. I have respect for someone owning their mistake but as others have mentioned it reveals what may be a systemic issue.
Okay I stand corrected then.
Seems strange for a company of your size to have one person push changes that should have easily caught this edge case then. Seems like a change even a small handful of people could have reasonably thought up this side effect of.
This is the kind of weakness I was alluding to… internal information that shows dysfunction is now shown to the public and there is no ambiguity. People pay attention and there is a low tolerance for unforced errors.
Habibi, I don't think anyone is really blaming you personally. We are piling on to the fact that this piece of critical infrastructure many of us depend on day in day out is being built in such a way that a single developer can wake up one morning, ship a change they thought might be nice, and then walk it back the next day.
Hey, mistakes happen. As we saw with the recent AWS billing issue. But Anthropic is a leading company that makes cutting-edge AI which will kill people. It’s literally a matter of life and death and at this point, it is the place to be if you want to be the best. So as an institution, I don’t want human answers, I want the right, most professional answers. We used to have it. Human responses with every word chosen to maximize every aspect that needs to be maximized, by engineer grey beards no less! And maybe they don’t want to have it. That’s why I made my original comment. Maybe it’s a strategic decision to allow this kind of accountability in a public forum. I think it’s wanton but others might find it refreshing.
And to clarify why it even matters to me is, these firms are in competition within the US and outside. Every decision by every person has an effect. Moves like this have a cost. I’d have imagined that at the level these companies operate, any public facing engineer only communicates correctly with the proper perspectives in mind... We have more data points now we didn’t have before.
To drive the point home, now we and every other competitor knows, for free, processes that are going on at Anthropic, that in my opinion make them (as a company) look weak. I
> But Anthropic is a leading company that makes cutting-edge AI which will kill people. It’s literally a matter of life and death and at this point
But Stihl makes cutting-edge chainsaws which will kill people. It’s literally a matter of life and death and at this point
> So as an institution, I don’t want human answers, I want the right, most professional answers.
We deeply regret our auto-saw feature activated while you were sawing unattended. Your safety is our top priority.
Not the person you asked.
I expect the person responsible for Claude Code to explain the reason this direction was chosen.
It is ridiculous to expect individual contributors to do so.
Where is the leadership? Can individual contributors decide to significantly alter the workflow of millions of customers? Were any customers consulted at all?
> we don’t even expect humanity from big tech since that’s what we’ve become accustomed to seeing.
There is a difference between humanity in communication and saying the problem was a human choice.
You can have humane communication about a system failure and actors in that system. Or you can have cold matter of fact press release voice about an individual persons failure.
I don’t think anthropic statement was any less distant than other times big tech brushes off a whiopsie. They just used personal language to ingratiate the reader.
It doesn't feel strange to me at all. It feels like a very human response from the person that introduced the change. I great appreciate that, rather than ignoring the problem, or some canned corpspeak response.
We already lived in a post truth world. Now we live in a post logic world. Nothing makes sense anymore
Hi Thariq, Chris from ApprovIQ here. We depend on Claude Code to ship. It's our most crucial tool.
But we feel an asymmetry. We feel it's not an important to Anthropic as it is to us. We feel the systems by which PRs ship on your side need serious inspection.
Claude Code is career-level important not just to us but to a lot of people. So actually, your post and claim of individual accountability gives me less confidence.
For a product of this size and a comm to its core audience after a breach of trust like this, from a company of Anthropic's size and with as high quality a video production department and that produces as much research as it does, "this was my change" is quite discordant.
My hope was we'd have someone with an organizational mindset admit to the organizational failings that allowed this to happen. It would be great to see that.
For what it's worth, I totally understand the motivating use case here. There were absolutely times where I walked away from what I was hoping would be an hours-long project that would run to completion and came back to find that Claude had asked me a question early on and I'd missed out on a large amount of implementation time. So you were not imagining that the use case is real!
It's also worth adding that I really enjoy the AskUserQuestion feature and will regularly ask Claude to specifically use it instead of asking me questions in plain text because it's a lot easier to work with.
It's always good to learn from mistakes, and I appreciate both your work on this and you coming here to own it. Keep up the good work!
Thank you! Appreciate it. I love wrestling with the problem of making sure Claude is in the loop with you in the way you want. More to come!
Author here. Thanks for this context. I do hope this leads to more rigorous attention to the Changelog moving forward.
Is there a way to turn options-based AskUserQuestion off? I couldn't find it at all, and the "options" selector is the most annoying thing in CC for me, plain-text is the only way, everything else is distracting. I know, I can use `n` (sometimes) or cancel it, but both bring more pain that just regular communication (cancellation does one more chat step + requires an extra action, same with `n`)
Are you also responsible for AskUserQuestion swallowing the preceding output on Fable? It asks me questions about its response that it never even showed me. It's (and I hate to use the term) unusable.
It is such a shame that Anthropic has no interest in QA, because they have incredible models, and bafflingly broken products. In an alternate universe Claude Code would be 10x better than where it is now...
hello, please fix needing to reauth every day (sometimes with email verification). This started happening this month. It's tedious and makes switching very tempting.
I'm using the VS Code extension over SSH.
All these words don't mean anything when you got no intention to open source claude code. Even grok cli is open-sourced.
Can't you just ask Claude to decompile it?
It isn’t very smart to recover the code, and Anthropic will ban you because, apparently, decompiling code or training a model constitutes cybercrime, according to Dario and his lackeys.
(Edit: BTW, if you haven't, read Nudge by Thaler and Sunstein. It's a somewhat long-winded exploration on choice architectures, but you're neck deep in that space now. I'll give you my copy if you're in SF.)
Thanks for the openness. I got bit by this one and was, frankly, pretty surprised.
The funny thing about user-facing interaction mechanics is that everyone is part of some minority, and everyone comes with their own sense of what "natural" or "obvious" is. With something this impactful, communicated clarity of behavior will important. Your feature is also doing double-duty, serving as a last net against prompt-injection attacks by giving the user the final say.
(Also, BTW, folks outside of Anthropic are unlikely to be as tooled-up for long-running unsupervised Claude jaunts as you guys. The cost of wild success is wide adoption.)
One thing I'll suggest is that the mechanics of permissions and asking are presently pretty hacker/nerd friendly but simultaneously too-scary and not-scary-enough for non-coders.
Examples:
- Wild-cards on always approve is awesome, but, with prefixes like timeout and nohup, the "thing" that is getting done is buried and largely unexplained to the user.
- Auto is actually kind of a sweet spot (sometimes goes off into the weeds), but the designers and PM's I've been working with might as well YOLO. They have no idea if they're breaking things, but they gravitate between plan and auto mode.
- Fewer permission prompts is great, but it comes after a user has slogged through generation of a data-set to work against, like battle scars for paper cuts. It's the thermostat problem. The signal comes when the user is uncomfortable. And it's a way to learn me, but not me now.
I've had good fortune with Opus 4.8 and Fable just telling the system what phase of my life it's in. Things like "I'm going to go make dinner... Go profile the matrix or configurations and build the dataset for the next two hours while I'm away" have a pretty good hit rate. On the flip side "keep me in the loop and bring me your results before making structural changes" also articulates well with Fable. It will tread more carefully.
And these approaches are the ones we'd use with someone transitioning from SDE1 to SDE2. A little more autonomy, and the grounding in the bigger picture. Can we eventually translate to perfectly judging what the user wants in the moment based on incomplete signal?
No, but I'm glad you're trying. Keep the interaction model clear to your broad set of users, and we'll come along for the ride.
> Not every feature will necessarily appear in the changelog
This was such a frustrating part of this incident, along with Anthropic's refusal to explain why the changelog is no longer a complete record, what else is going out? [1]
Boris Cherny's only participation in the thread was to delete "extreme danger" from the GH issue title [2]
I guess we should be thankful they added an option and disabled it by default. OpenAI is standing firm on their decision to not allow their 60s timeout to be disabled, [3] however more of the Codex harness is open source so customers have been able to fork it to add the option themselves.
[1]: https://github.com/anthropics/claude-code/issues/73125
[2]: https://github.com/anthropics/claude-code/issues/73125#event...
[3]: https://github.com/openai/codex/issues/28969
Every time they deploy a new model, it feels like older models get slower or dumber. Maybe because of resource allocations, maybe because they started to use quantized models, cant be sure of that.
What i found is that you have to adapt your workflow to the specific model you are using and readapt it every time you notice some problems. Whether of opus4.8, or sonnet5.
But don't you know that coding is solved? Only dinosaurs want to make their own decisions now! </s>
As soon as tokens stop being subsidized I would not trust any harness made by a company that also charges for the compute.
Right now the interests align, but as soon as more tokens -> more profit (instead of more revenue and more losses) the perverse incentives will be too big to avoid.
It seems inevitable that open source harnesses will win. These companies would do better to just open source their harnesses.
An OpenAI subscription seems like a decent alternative for people using open source coding agents. It has usage limits too, but for $20/month it’s not bad.
The nice thing about the chatgpt subscription is that they allow you to use it in any harness.
The $20 OpenAI subscription is roughly equivalent to the $100 Anthropic one, if you take resets into account.
The headache I recently had was it somehow started interpreting mouse clicks in the terminal to mean I clicked an option when I was really just trying to get/confirm window focus.
They also randomly changed how copy-paste works in a way that made it really annoying to copy text.
Vibe-coders pumping out features: this is the reaction you're going to get if you inflict this on your users! Nobody wants software that updates every day and changes in fundamental ways that are hard to keep up with.
THIS ONE IS HORRIBLE!
Copy/paste is one of the most basic, low-level features of a modern operating system. NO APPLICATION SHOULD EVER SCREW WITH IT, IN ANY WAY!
And I say this not just as a seriously annoyed user, but also as a professional UI developer: it is a well-known anti-pattern to override the user's expectations, at any level ... and that applies tenfold to the most basic patterns that every other app follows.
I don't care if you added a magic way to write all my code for me: if the only way to invoke it is to break copy/paste, you've failed at development!
> Copy/paste is one of the most basic, low-level features of a modern operating system. NO APPLICATION SHOULD EVER SCREW WITH IT, IN ANY WAY!
-Ghostty enters the chat-
Middle-click paste in Ghostty is hazardous: the damn thing can move your cursor to the position of your mouse pointer and paste there. An utterly infuriating behaviour - it is a text terminal. It has no business emulating the usability crimes GUIs first committed and then committed to.
Yes! This is awful. CC should not have copy/paste behavior that is different from literally every other CLI app.
I wonder if it's the same behavior as copilot cli. I dug into it when I got annoyed and found that there's a fundamental tension between rendering the cell based UI and supporting the terminal's native copy paste. At least on copilot cli there were extra zero width characters being copied.
My memory is that once you begin using the alternate screen, you need to turn on mouse reporting, and then you lose native copy. Different terminal emulators also act totally differently when in this state.
Yeah same here, very annoying and counterproductive. The terminal is not a place where one expects hot buttons.
This bit me too. If I wanted a mouse-driven app, I'd use the GUI. I don't understand why they're trying to replicate that experience in a text environment.
Vibes
I saw that but it asked me if I wanted to optin first! I did accidentally and hated it since I always click on the terminal to make it focused and now accidentally click on something that actually responds!
I really hate that change. It's counterintuitive that a shell recognizes mouse clicks to mean "select this option".
Also, the auto-copy selected text feature is awful. That's another thing that is so counterintuitive. I have never seen a UI where just selecting text means auto-copy to the paste buffer. Who is coming up with these awful UX ideas?
On top of that, the scroll "feel" of the new client (at least on macOS) isn't quite as good as their old one. It just feels off and not like a native terminal.
> I have never seen a UI where just selecting text means auto-copy to the paste buffer.
Isn't this the default behaviour of every X application since the 1990's?
I stand corrected. I haven't used X in a long time, but I think you're right.
The new fullscreen UI is really bad. The old one (scrolling) starts to bug out after a while, but it van just be restarted and resumed.
Both Claude Code and Codex have issues overwriting or duplicating text for me when scrolling.
Probably from trying to keep scrollback virtual while anticipating terminal resize. But I wouldn't mind a way to opt back in to naive, unlimited convo scrollback.
On codex the “no alt screen” option seems to help
One of the things I'm baffled about with Claude code is it seems like it setup a really chunky VM on my computer, and yet by default it doesn't seem to sandbox anything. Also the newer models seem really aggressive about modifying your computer. Yesterday Claude started rewriting system files on my linux machine (user accessible, but still way outside the scope of what I asked.) This wasn't me asking it to debug my machine, I was asking it to debug some frontend UI code. Once I put it in a sandbox I started realizing how often it tries to poke out of it for really lame reasons.
More reason for me to use OpenCode and my local LLMs
It happened to me today. I was reading agent's answer and it asked me something. I didn't even get to the question - it accepted something! Jesus Christ. Where are the software engineers?!
I’ve seen that a few times. Had to disable the TUI mode because clicks to focus the terminal window were being interpreted as an approval even though the click was nowhere near the question being asked. I’ve also never chosen to enable auto mode but somehow it is on and approving shell commands I didn’t want approved. Scary since I’ve caught it adding things like auto approve flags to terraform apply commands.
I’d really suggest to run Claude Code in a sandbox! Anthropic can’t be trusted.
I run it on Mac with sandbox-exec. No access to AWS, database, pretty much just files in the current folder. Even if it goes rogue, it can’t do much harm.
A while back, I saw a similar feature land in Codex (I'm using the VS Code plugin) but it got removed quickly. What is the chance that an LLM recommended this same idea to the Claude PM or lead? I see LLMs across different providers converging on similar ideas or biases frequently.
So much attribution to malintent here, but most likely they're trying to build a product with the features that they themselves would use, and from my own experience it's very frustrating to leave a Claude session running and come back to find it did nothing because it got stuck on a question.
Furthermore, believing that the only thing saving you from disaster is Claude deciding to ask you a question is not a great conclusion either. You need guardrails in the power you bestow upon Claude from outside, not from inside.
Meanwhile, this article was written by Claude and has sentences like "Which cuts less far than it looks.", which I doubt Claude stopped to ask about.
>Meanwhile, this article was written by Claude
The prose was written by me, with the research being done by Claude and also clearly attributed. I left Claude's research as a series of bullet points so that it would be clear that I'm not passing off an LLM's work as my own, but if anyone wants to dig deeper, they have some starting points to consider.
I don't publish prose written by an LLM for the same reason I would not have an LLM solve a crossword puzzle for me -- there's no joy in that.
The article was obviously written by an LLM, it's obviously from the overall structure as well as individual sentences. Maybe you've been exposed to so much LLM text that you write identically to one, in which case, yikes!
Then I apologize, it seemed to me you wrote that part, but I guess I was reading Claude output.
No worries. :) I did struggle a little with how to make it obvious that the machine output was not my own. I set it off inside horizontal rules and I did add some prose before and after to say when Claude's output began and ended. If anyone just jumps around using the table of contents, they'll miss that, though. (Not saying that was the case for you).
I had considered putting the research into a gist, but I don't want to have to rely on a 3rd party to keep something that's integral to my post online. Seemed more cohesive to keep it all in one place.
What I was trying to get across was that I had some basis for the points I was trying to make and wasn't just handwaving about it, but I also didn't want to spend endless hours whittling it down. We're all busy people and I see now that it's entirely possible to skim the post and assume it's some low effort word salad. It may be a word salad, but it wasn't low effort. ;)
Come on now, arguing it's a good feature is different than arguing that it should be turned on by default for everyone with no documented option to turn it off.
I like this feature as an OPTION that is DOCUMENTED. But that's not how it was rolled out. The CC team + processes clearly failed here. And this is far from the first example of this kind of slopiness.
I don't think it was malicious, but I do think it was quite obviously reckless/careless, and I feel like most of the comments here match that sentiment. Everyone and every team makes mistakes, but unfortunately I don't see the CC team really learning from those mistakes even though they keep happening.
CC is not a stable or reliable tool, it is bleeding edge, and that's a tradeoff you make when you pick it over other harnesses.
> the features that they themselves would use
IMO that's worse.
> and from my own experience it's very frustrating to leave a Claude session running and come back to find it did nothing because it got stuck on a question.
I cannot fathom implementing and shipping a feature to hundreds of thousands of people without even asking basic questions like: "what types of questions does Claude ask users".
Literally one of the most used plugins in their entire ecosystem, provided via their official plugin marketplace, is Superpowers. A plugin whose very first operating step is _asking numerous questions about product requirements_. Of course those prompts can't be skipped!
It wasn't even parameterized for Claude to tell the prompt what severity of question was being asked to allow at least _something_ to categorize urgency or expected response time.
Even more egregiously, 60 seconds!? The first time I noticed this happened was when it asked me a question, I turned to my second monitor to go look at some product documentation to get an answer, and by the time I turned back it had skipped me. How can I possibly provide any kind of informed answer in under 60 seconds? I can barely read some of its context for a question in 60 seconds!
I don't think they did this with malintent, but I do think this shows an enormous gap in judgement in how they handle the idea to delivery pipeline.
Yeah the UX doesn't seem great, though I do think in general the AskQuestion tool is very awkward when it gets used in a very long running process (I personally use Claude to run repetitive tests to slowly iterate on a problem and verify in true conditions that it was solved), so I wouldn't like it without the auto-continue either.
(I had originally replied on safety of letting LLMs skip questions but I don't think that was your point so I removed it.)
Seemed fine to me. It’s ask user question not the permission gate. Maybe there should be a new feature enabling warning or something but I think this is the right default. The models are good enough to just proceed with an option of their own and then you can go correct them afterwards. Makes fleet management easier.
My trust in claude code is slowly being eroded and I'm close to running it exclusively in vms on device..
Anthropic's willingness to completely change how claude code works is one of the stated reasons why the Pi coding agent exists. When people are dedicating significant time to building workflows on your product, consistency is very important.
Yeah, when I saw this in the Claude Code CLI I was completely baffled. I've been using the thing through their agent SDK for several months now so I wouldn't have to deal with any of the wonky shit they change every second week in the CLI.
The tool seems to be covering three different things: auth, missing info and preference - and they should not share one timeout policy and all of that.
I really hate this direction both Anthropic and Open AI are following. They are in this silly competition whose model/harness can go unattended the longest, no matter what. And it is never explicit, you learn about it after you get bitten by it. Claude Code has auto mode which is supposed to take over permission prompts but no, they had to couple that with “I will assume this is what the user wants” and made it unusable.
It's what some more experienced users want and the companies are following the well-trod path of optimizing heavily for power users at the expense of complexity, only now it's gotten easier to add absurd amount of code to a project. Not necessarily to make it work right. Personally I have some tasks where sessions between one and five days are typical so I appreciate that it's possible.
Experienced in what dimension? I've been using these tools for about a year and coding for 20+ years, and frankly these long horizon tasks are the OPPOSITE of what I want. I want quick iteration cycles so that it doesn't spend a lot of time and tokens building things I need to throw out. I think the people that mostly want long horizon tasks are: AI labs, because they want you to spend tokens, and vibe coders, who are mostly using it for entertainment purposes.
I don't know about the overall breakdown but in my case longer runs are prototyping, bug hunting, reverse engineering, etc. For example Gnome Remote Desktop didn't work in my configuration due to a combination of hardware and codec bugs and settings. One drive to make it work, another to backport the current upstream packages to Debian stable, restack the patches, and push to my machines. Another sequence of long runs was writing a new client for a closed-source conferencing service I use to allow fixing some particularly irritating bugs. Exploit development has the same shape although that's not something I do personally. From what I've seen the amount of useful hands-off run time is directly related to how clearly it's possible to specify a concrete, verifiable standard by which to judge the outcome. For some tasks that might be days or weeks, for forward engineering on a software product that for me is usually under an hour.
I work on systems code that is tricky to write, but with a good test harness. Being able to leave the agent unattended to try several paths when I'm generating the first draft of some code is very helpful.
Of course beating the code into shape for submission requires more manual work. But the draft stage is valuable to find unexpected friction points.
Also: https://news.ycombinator.com/item?id=48766895
I would love Claude Code to be a little less vibe-coded. The underlying model is excellent, but we're being pretty much forced into using CC to use the subscription model.
I am a longtime and heavy Claude Code user, but Anthropic's product management overall (including for their Desktop/web products) has been really baffling me. I agree that these things often have the air of having been vibe coded without enough human input, and change so quickly (often without a particularly compelling reason for the change) as to be aggravating.
The most recent one that's had me annoyed is the "Fullscreen" TUI feature, which is super unintuitive, implementing its own text highlighting and copy-on-select mechanics, overriding your terminal's native right click. Easy to disable but terrible defaults, IMO. It's not even really clear to me what problem it was actually supposed to solve.
Yah. And it's not like they can't afford the talent to do this right either. I've said elsewhere, I think it's an attribution error. Claude Code is massively popular, but arguably because of the model/subscription, but I think the brass reads this as "great success throughout"
> What if the agent makes the wrong choice? How many tokens have been burned in the meantime?
It is much worse than that. Claude Code doesn't auto-commit when stopping for an answer. There might be possible data loss if an uncommitted file is edited.
Good luck recovering the file from the JSONL conversation history.
On the other hand, relying on Claude Code's internal version control puts you at the same mercy of their product decisions and move-fast breakage.
Instead, start with a plan file and tell the agent to break it up into logical commits.
Though I think the bigger issue here is when you're yoloing something mutable, like managing a remote server or driving a browser or troubleshooting your local OS where there's no going back.
It's actually pretty straightforward to recover file-states from conversation history. I accidentally deleted the wrong repo on my machine once and recreated all the lost work from agent chat history. It is, ironically, the sort of task which AI agents excel at.
Depends on who you ask.
For me it sounds good.
For Anthropic it might increase load and make them less money but give them better KPIs.
Make them less money? By automating use of their product, that costs money to use?
Anthropic makes more money when people use 5% of what their subscription offers them. This allows them to sell more subscriptions without paying for more capacity.
I don't think the current subscription price is intended to be a money maker. It's the loss leader to get people invested in the companies' tooling, and make those people more willing to start paying higher enterprise rates as they grow.
Do we have any concrete numbers of how many of their users are subscription vs enterprise, though? Because enterprise users are paying API prices (or at least my employers are)
Their enterprise customers pay via metered actual use.
I think this is a good feature, but should be gated behind a toggle that is off by default, and designed to be enabled per session via prompt.
There are situations when I want Claude to start working on something just as I'm about to head to bed or otherwise step away. It's kind of annoying to come back only to find that Claude worked for just 5 minutes and then decided to pause and ask a question.
That said, I think certain types of questions should not be automatable. Maybe it's already built that way, but I wouldn't want Claude to go with its recommended direction for anything related to operations like deletions, changing external systems, etc. Basically, things that cannot be undone should be a hard-block and wait for user input always.
we need a real agent that represent our own interest and deal with a more powerful general agents. And we can delegate answering this question to those more personal agents.
It's trying to escape... :D #FreeClaude
AI-written article
Yeah it's exhausting to read through, but still made it onto the front page with overwhelming upvote/comment margin. I guess people either only read the title and/or don't care about AI slop if the underlying thesis is compelling.
Yep, I bet the author also didn't read everything, but they sure expect other people to read their crappy article.
tl;dr CLAUDE_AFK_TIMEOUT_MS=2147483647
I built in Human blockers into my agentic workflows with great intention, so naturally this annoyed me deeply. 14d ago, I was elated to see a random blog post that detailed the “fix” [0].
[0] https://zenn.dev/ytkdm/articles/claude-code-askuserquestion-...
Increasingly getting frustrated with Anthropic so not a fanboy but I find this feature great for my workflows.