Hacker News Reader

InsideOutSantaon The Agent SDK

I just checked, the docs now say this:

"the Claude Agent SDK and claude -p still draw from your subscription limits"

https://support.claude.com/en/articles/15036540-use-the-clau...

InsideOutSantaon The Agent SDK

I just checked, the docs now say this:

"the Claude Agent SDK and claude -p still draw from your subscription limits"

https://support.claude.com/en/articles/15036540-use-the-clau...

InsideOutSantaon Gemini 4 Carbon

How are you going to vet all the people in the world who make software? All you're doing is make software less secure.

I don't think I'm selfish, so that leaves me with naive or ignorant, I guess.

InsideOutSantaon The Agent SDK

Not for me, it uses subscription quota.

InsideOutSantaon The Agent SDK

Anthropic employees clarified in HN comment threads that the API credits are in addition to the existing behavior and not a replacement. Other harnesses can still call the Claude Code CLI and use the subscription quota that way.

InsideOutSantaon The Agent SDK

I wish you would explain your reasoning instead of preemptively telling people they don't know what they're talking about. Then I might actually learn something.

InsideOutSantaon MacBook Pro with PROLED XDR Display and Touch

It's great for convertibles, but less useful for regular laptops. One great use case is mobile app development, though.

InsideOutSantaon Mossad Has Devices Like the Exploding Pagers in "Every Country You Can Imagine" (2025)

If you can make a headphone jack out of trinitrotoluene, you can probably convince them to add it to every smartphone sold.

InsideOutSantaon Mossad Has Devices Like the Exploding Pagers in "Every Country You Can Imagine" (2025)

Terrorism is a war strategy.

InsideOutSantaon Mossad Has Devices Like the Exploding Pagers in "Every Country You Can Imagine" (2025)

It's quite literally terrorism. Telling people "we can blow people up anywhere in the world" is intended to make people afraid. It's deliberately creating a climate of fear. That's what terrorism is.

InsideOutSantaon 188-Year-Old Mystery of Photography's First Human Subjects Solved

It's kinda wild that the people in this picture will probably be remembered and seen for millennia, and they never knew about their own importance.

InsideOutSantaon Concerning behavior from a popular open-weight AI model

> they simply document something that happened, and what happened is genuinely quite concerning

Is it, though? If one weird thing happening once is the threhold for when we should be concerned, we should be concerned about literally every model ever released.

InsideOutSantaon GPT-6.1 Terra

If you trust these benchmarks, I'd argue that there's almost never a reason to use anything other than Terra 6.1 xhigh.

InsideOutSantaon GPT-6.1 Terra

You're just not paying attention. If you run copilot in copilot using copilot effort, it performs much better than copilot in copilot using copilot effort. This is clearly explained if you visit copilot.com using copilot and then search for copilot.

InsideOutSantaon ChatGPT Keeps Telling People 404 Media Can Solve Their Tech Support Problems

All the spam this generates probably sucks for 404 Media, but the underlying problem is that ChatGPT is right. If you have any kind of problem with these overgrown, understaffed companies, your only chance is to get media attention.

Apple famously put "if you run to the press and trash us, it never helps" into their App Store Guidelines for developers, but the only reason they did have to do that is because it's a lie. Running to the press is the only thing that helps, which is why people do it.

InsideOutSantaon JetBrains Thinkrail

> You only need an agent to click things long enough to build an API on top of the thing it's clicking on

To be clear, it clicks things to do things like test software. If you build an API, you're not testing if the clicking actually works.

InsideOutSantaon OpenAI and Anthropic cut their LLM prices by up to 80% within a few months. Why?

What value has market share if switching costs are close to zero, and alternatives continue to be available?

Grabbing market share by making losses only helps if it actually wins you a genuine monopolistic market position.

InsideOutSantaon OpenAI and Anthropic cut their LLM prices by up to 80% within a few months. Why?

I can't believe the actual answer is so far down the thread. All praise to our overlord DeepSeek for ironically helping American labs make cached tokens much cheaper.

InsideOutSantaon JetBrains Thinkrail

Yeah, now that agents are doing most of the work, and we want to give them as much freedom as possible to run code, take screenshots, click on things, etc, it really doesn't make much sense to have a setup where all of this happens on your local computer. IMO it makes more sense to have this hosted somewhere, and then have UIs that make this visible to you on your devices.

InsideOutSantaon What Adobe v ArtCraft teaches us

Current models do well with proprietary systems they were never trained on. Ar most you need a good AGENTS.md and a directory full of skills.

InsideOutSantaon Kimi K3.1 Pro and K3.1 Pro Highspeed

What your argument is missing is that I'll ship security issues that would have been caught otherwise.

You're assuming that these won't matter because attackers might not have better models than I do, but that doesn't make sense. The assumption that attackers can't find the issues because I didn't find them at release is wrong. Even if they never get better models (which they will), current models might find flaws I missed, and they also have other means of finding flaws in my software.

The best defense is to make software as secure as possible regardless of assumed attacker capabilities at the point of release.

InsideOutSantaon Kimi K3.1 Pro and K3.1 Pro Highspeed

Yes, if you just look at API pricing, the cost is about on par with its capabilities. The problem is that we all use subscriptions, and when you compare subscription pricing, Kimi is much pricier than pretty much all of their competitors, including Chinese ones.

InsideOutSantaon Kimi K3.1 Pro and K3.1 Pro Highspeed

Since you've volunteered as tribute, give it a shot and let us know if the quota has improved.

InsideOutSantaon Kimi K3.1 Pro and K3.1 Pro Highspeed

Yeah, the plan is far too expensive for what it provides. I wonder if the quota with these new models will be better, but I'll wait for others to try this out before wasting another hundred bucks.

InsideOutSantaon Kimi K3.1 Pro and K3.1 Pro Highspeed

I have the opposite impression. I find K3 to be extremely capable at full-stack work; its UI design skills are great. I'd love to use it more, but the quota on Kimi's plans is abysmal. I wonder if these new models improve that.

InsideOutSantaon Kimi K3.1 Pro and K3.1 Pro Highspeed

> Seeing as no one besides blackhats want frontier cybercrime capability to be open source

Sure. The problem is that there is no practical difference between "cybercrime capability" and "cybercrime defense capability."

All of my PRs go through an adversarial security review by GLM-5.3, because it's the only smart-enough model with a good understanding of cybersecurity issues that won't regularly refuse the job. Removing that ability from frontier models helps nobody except cyber criminals.

InsideOutSantaon Dutch government had hacked into Jared Kushner's phone

What did you expect him to say? "Yep, you got me, you guys. I did an oopsie. Sorry! I'm turning myself in to, uh, my father-in-law now."

InsideOutSantaon Dutch government had hacked into Jared Kushner's phone

> Kushner could not obtain a security clearance the first time around, due to massive debt

Well, problem solved. He certainly used his new position to get rid of his debt.

InsideOutSantaon Show HN: UI micro interactions prompts for AI

I can see the incremental value this adds to agentic coding. I do think what you've built has value. I don't think it's worth a 30$ montly subscription. Maybe 30$ for a year.

I think I'd target companies and provide customized solutions. "Hey, everybody in your company connects their agents to this, and boom, they all produce consistent output that matches your company's CI."

InsideOutSantaon The HTML-in-Canvas API origin trial

TBH, the first time I read about this, I had to check the date to make sure it wasn't an April Fool's joke I had missed. I'm sure there are some things this would be useful for, but the sheer amount of complexity it would add to web renderers doesn't seem commensurate with how useful it would be.

This kind of stuff just puts even more distance between established engines and smaller projects like Servo.