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

Arn't they yoinking an OAuth token for replay in the Claw app?

If so, I don't think anybody who knows how auth works could feign complete innocence.


People got banned for calling `gemini -p` (non-interactive mode) from wrappers like pi or opencode, too.

I understand how grabbing an oauth token via reverse engineering could be a ToS violation. But there's no other purpose for the `-p` flag other than to use it with a wrapper. Unless people enjoy having interactive conversations via non-interactive mode for some reason.

Even their documentation clearly states this flag exists for "building custom AI tools" [1]. How is OpenCode, OpenClaw etc not a "custom AI tool", where exactly is the line drawn?

This is just a rug pull.

[1] https://github.com/google-gemini/gemini-cli/blob/c7237f0c795...


Two kinds of oauth tokens at play here.

Tokens like those generated by the gemini CLI, accounted to the user and metered so you're charged every million or so tokens.

Tokens like those internal to the Antigravity app, accounted to the product (which is why this was noticed -- the resources allotted to the app were being exhausted) and not metered per token (which is why everyone was trying to use them for everything).

As the other commenter said, the first person to do this definitely knew what they were doing. All the script kiddies who piled on after, probably not, but that's also part of why "script kiddy" is a derogatory term. Not only do they usually not understand the scripts they run, they also typically do not understand the risks associated with them.


I don't know if you are just playing devil's advocate, but there's plenty of examples of code quality issues coming out of msft these days too.

Regardless, their point is that the argument seems faulty. Indeed, their docs going unreviewed seems moot to whether the code goes unreviewed, given there are much stronger reasons to review code than there are to review documentation; as they wrote, bad documentation doesn't automatically break your application when it's published (there's at least a few more steps involved). Your statement's accuracy is not exclusive to the illogic of an argument which agrees with the statement.

> I don't know if you are just playing devil's advocate

Indeed, that is playing Devil's Advocate but one should remember that such Advocacy is performed to make sure that arguments against the Devil are as strong as they can be. It's not straightforward to see how simply repeating an assertion helps to argue for the veracity of it.


They never said that there was no example of code quality issues.

What they say is that low quality in the documentation does not mean low quality in the code. Nothing says that they are related.


Sure I suppose on the face of it the two are not necessarily related, but it seems likely to be highly correlated?

> these days

I realize BSOD is no longer nearly as common as it once was, but let's not forget that Windows used to be very fragile indeed.


It was more fragile 20 years ago than it is today.

It was more robust 5 years ago than it is today.

Or at least that's been my impression. I can't back that up with hard data.


>> I realize BSOD is no longer nearly as common as it once was

Anecdotally, installing wrong drivers (in my case it was drivers for COM-port STM32 interaction) could make it as common as twice a day on Win11. While my windows server 2008 still doing just great, no BSOD through lifetime.

I agree that for a common user BSOD is now less likely to happen, but wonder whether it's less to do with windows core, and more with windows defender default aggressive settings


I'm building an agent wrapper over Claude Code, and use the "Jobs done" peon voice for notification (there are two variants).

For when user attention is needed, I play a few seconds of Rick Astley's "Never Gonna Give You Up". =D


if you enable verbose mode, it does.

However I run like 3x concurrent sessions that do multiple compacts throughout, for like 8hrs/day, and I go through a 20x subscription in about 1/2 week. So I'm extremely skeptical of these negative claims.

Edit: However I stay on top of my prompting efficiency, maybe doing some incredibly wasteful task is... wasteful?


this is about variance of daily statistics, so I think the suggestions are entirely appropriate in this context.


their c# LSP theoretically worked for a week or so (I never saw it in action though), but now it always errors on launch :(


There was an issue in Claude code which is fixed in latest release


what's the make/model for the monitors? my setup is getting long in the tooth.


Two PA27JCV and one LG ultrasharp (it was cheap because it was broken and I repaired it) and the 4k monitor is a samsung which I cant recommend. (Open box was cheap though)


I bet the friend just pressed the Windows key, and typing "Calc" and quickly pressing enter caused Bing to search for calc instead. Common failure because window's start-menu search/load/discovery is a total mess.


Even in this case it opens the calculator. Web search results are further down.


if you are searching for something for the first time (or after caching invalidates), it seems like it prioritizes search sources that have already completed.

on my computer, that means web-search almost always completes first. So most of the time if I type in something "new" and don't wait, it'll bring up Bing.

Sometimes it looks like "downloads folder" file search completes before Installed app search completes, because on one occasion I typed in an app's name and it launched the INSTALLER for the app.

once all the searchs resolve it behaves "as expected". I am very surprised if you don't have the same symptoms I'm describing. Why is your computer behaving different from every Win11 install I ever interact with?


I just tried a search for "downloads" and the first result was "Downloads folder privacy settings". I never search for that so it wasn't cached. I even pasted in the query to give it less time to search before pressing enter.

I don't think I've changed any settings for search. Everything is still enabled. There's over 250,000 items in the search index so I haven't removed indexed locations. My computer is pretty much a high-end gaming PC using last generation CPU and GPU. But really I've never seen this behavior anywhere - including my very basic laptop. Maybe I could see this happening on computers that are still using a HDD but I haven't tested that.


pretty weird, i have a few moderately high-end pc's and cheap laptops and they all have the same issues. Maybe me disabling a bunch of telemetry stuff screws up the caching.


I don't think caching makes the difference for me. I never search for "downloads" so it shouldn't be cached. Calculator probably was cached.


I see you were fortunate enough to not use notepad aprox 5 months ago, when they were running the rich formatting preview. It was on by default, and would drop around 5% of the characters you type. Literally failing at the only thing it's supposed to do. I repro'd this on 2 out of 2 machines.

Maybe they fixed it, maybe they haven't. I both turned off formatting and am using vscode for notes now.


I think I had to disable spellcheck to fix the ignored keystrokes, it happened even after disabling formatting


ahh, it might have been spellcheck then. I turned off all that stuff. In the heat of the moment, maybe I was a bit too angry to do proper root cause analysis :P


Good info. Now I understand why they refused to acknowledge the UX issue behind my bug report: https://github.com/anthropics/claude-code/issues/7988

---

(that it's a big pile of spaghetti that can't be improved without breaking uncountable dependencies)


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

Search: