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

I've had a lot of luck with pyghidra-mcp -- give it a try :)

Well i have tried and it only works for simple use-case.

I live in a walkable city. I cannot drive because I am blind. Cars make my life better. Uber exists. I use it to get many places that I otherwise wouldn't go to.

This doesn’t contradict or respond to the comment you are replying to in any way.

1.35 times! For Input! For what kinds of tokens precisely? Programming? Unicode? If they seriously increased token usage by 35% for typical tasks this is gonna be rough.

Take boots, for example. He earned thirty-eight dollars a month plus allowances. A really good pair of leather boots cost fifty dollars. But an affordable pair of boots, which were sort of OK for a season or two and then leaked like hell when the cardboard gave out, cost about ten dollars.

Those were the kind of boots Vimes always bought, and wore until the soles were so thin that he could tell where he was in Ankh-Morpork on a foggy night by the feel of the cobbles.

But the thing was that good boots lasted for years and years. A man who could afford fifty dollars had a pair of boots that’d still be keeping his feet dry in ten years’ time, while the poor man who could only afford cheap boots would have spent a hundred dollars on boots in the same time and would still have wet feet.


Maybe they should just cap the number of orders at the number of items they can make and ask anybody else to sign up on a list? Anybody who chooses option 1 is obviously evil?

100% this. If you can't deliver the product i want, then fine. Don't lie to me and deliver a product inferior to what i ordered for the same price without warning. That's straight up malice.

Naturally the kind of thing that would be defended on HN nonetheless


You'd think that if they were compute-limited ... Trying to get people to use it less ... The rational thing to do would be to not ship features that will use more compute automatedly? Or does this use extra usage?

I would imagine that this sort of scheduling allows them to have more predictable loads, and they may be hoping that people will schedule some of their tasks in “off hours” to reduce daytime load.

It also beats OC's heartbeat where it auto-runs every 30 minutes and runs a bunch of prompts to see if it actually needed to run or not.

Man, this just bit me too. I started playing with OC over the weekend (in a VM), and the spend was INSANE even though I wasn't doing anything. I don't see this as very useful as an "assistant" that wanders around and anticipates my needs. But I do like the job system, and the ability to make skills, then run them on a schedule or in response to events. But when I looked into what it was doing behind my back, 48 times a day it was packaging up 20K tokens of silly context ("Be a good agent, be helpful, etc, for 30 paragraphs"), shipping it off to the model, and then responding with a single HEARTBEAT_OK.

Luckily you can turn if off pretty easily, but I don't know why it's on by default to begin with. I guess holdover from when people used it with a $20 subscription and didn't care.


I can recommend Hermes Agent as an alternative to OpenClaw which actually works well, is properly architected, and doesn’t break three times a week.

If you want something more lightweight, I made one that has no heartbeat by default: https://stavrobot.stavros.io.

It's very light on token usage in general, as well.


It’s really, really ridiculous just how many tokens OpenClaw burns when it’s not doing anything.

Also you can schedule it a bit off. Every hour? Delay it a few seconds. Can’t do that with a chat message. Also, batch up a bunch of them, maybe save some compute that way? Latency is not an issue.

I thought about that but I'm pretty sure that if the backlog is automatically clean and I don't need to run my skill for that when I start up in the morning that just means I can do the next task I would have done which will probably use Claude Code.

Your own, personal, Jevons.


They are more worried about building a moat than anything else. They want people building integrations that are difficult to undo so that they lock into the platform.

So basically what happened:

1. Anthropic realized their models weren't enough of a moat.

2. They built tools so they could expand their moat.

3. People don't want to use their tools, they want their models, and use other, better tools.

4. Anthropic bans the use of better tools, taking advantage of their model superiority to try to lock people into subpar tools.

"I don't have enough of a moat so I'll use my little moat and pretend it's a big one" doesn't sound like a great strategy. All they're doing with this anticonsumer behaviour is making sure that I'll leave the moment another model works for me as well as Claude does.


Kind of. Yes, they want you to use their tools.

I doubt Anthropic ever thought they would have a big moat just based on the models. The platform is just as important.

Claude code and Cowork would still be extremely valuable to Anthropic even if they didn't release them to the public.

Owning the harness gives them a ton of data they can use to tune the models.

This is a perfectly sane strategy even if it's a bit unsavoury to some technical folk.


> I'll leave the moment another model works for me as well as Claude does.

Isnt that a large moat in itself but you are claiming its not enough?


Apparently not for Anthropic, since they've been wanting to build a larger one by force. My point is you can't build a moat by forcing people, it defeats the purpose.

What are other better tools? I would like to use them

OpenCode, Pi, whatever Anthropic doesn't let you use with their subscription because they want to lock you in to their stuff.

> They want people building integrations that are difficult to undo so that they lock into the platform.

Ironically, they are now playing against their own models that can relatively easily build wrappers around any API shape into any other API shape.


Max accounts get 15 daily runs included, any runs above that will get billed as extra usage.

I don't think "usage" is exactly the metric they're going for, more like "usage in line with our developmental strategy." Transcripts of people using Claude to write code are probably far more valuable to them than transcripts of OpenClaw trying to set up a calendar invite.

I mean, they don’t train on your data unless you have the setting enabled. Do you really think they are reading your prompts at all? Free inference providers sure, but Anthropic?

It's how they can lock more users into their eco-system.

The rational thing to do would be to build more data centers. Use that huge influx of money on infrastructure. Make tokens dirt cheap.

So from the exact same article:

Doctors Will Never Be Ethical or Safe

Hardware Stores Will Never Be Ethical or Safe.

Okay?


I'm a cofounder of a company that builds screen reader equivalents for maps. What's the mechanism by which my role as executive harms humanity?

The mechanism of making it a for-profit product for the blind people who can pay for it, as opposed as giving it at cost, or even better, having it treated as a basic right of blind people, built and provided by a welfare state to any of them that need it for free.

How is selling products at a profit harmful? If the only reason is that you could provide it for "free" (*insert no free lunch*), then there's no actual harm. No one is entitled to labor/property of others.

On the contrary, however, a welfare state actively rellies on harm (through ~steali~, I mean taxes) to provide said "free" goods.


Brought to you from the same guy who said in April 2025: "Google Is Winning on Every AI Front" https://www.thealgorithmicbridge.com/p/google-is-winning-on-...

> It's weird that Aisle wrote this.

No, writing an advertisement is not weird. What's weird is that it's top of HN. Or really, no, this isn't weird either if you think about it -- people lookin for a gotcha "Oh see, that new model really isn't that good/it's surely hitting a wall/plateau any day now" upvoted it.


Nah, Saturday post. Less news less content.

It's not weird. Top of HN is worthless as a barometer at this point, people downvote for calling out AI slop.

Can you downvote submissions?

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

Search: