ghrl

6 ms·

1.
by ghrl 5d ago
Just what I thought. > made with <3 by Ayden That's the cherry on the cake - indicating genuine effort and attention to detail put into something, yet not even checking the credit at the bottom.
2.
by ghrl 1mo ago
I can't speak for Anthropic, but with Google's SynthID, the algorithm is actually public. However, checking (or creating) the watermark requires a symmetric key, and the providers likely wouldn't share that key.
3.
by ghrl 1mo ago
Yes, exactly, that is my understanding as well. Since the watermarking is based on a symmetric key by design (as to not be easily able to test and remove the watermarks), the providers will need to offer a watermark detection API. I see eve
4.
by ghrl 1mo ago
My biggest concern is that checking any text for watermarks requires sending the entire text to Anthropic. And even that is not sufficient, as the text might have been generated with ChatGPT, Gemini, Grok, Mistral, ... So every check requir
5.
by ghrl 2mo ago
From skimming through the article, isn't that what the uppercase variants do?
6.
by ghrl 2mo ago
What shines through here is that the AI writing / reasoning this is about as lazy as always, declaring a hard limitation where some engineering might solve or work around the limitations. In modern JavaScript a surprising amount of tas
7.
by ghrl 3mo ago
Beer CSS is great. I've used it for multiple simple projects and it provides a great DX with the clean html code and the many snippets on the official website. The only downside is that LLMs are quite bad at working with it from my exp
8.
by ghrl 3mo ago
That's refreshingly usable and not-slop-looking, nice additional style.
9.
by ghrl 3mo ago
Yeah, spot on. I had an agent delete some files it shouldn't have as well, similarly to me making the same mistake. I think system prompts should default to using `trash` over `rm`. For now that's just in my AGENTS.md, and gets ho
10.
by ghrl 3mo ago
Sounds like a case for NixOS
11.
by ghrl 3mo ago
Amazing observation, and I'm certainly guilty of it too, but it is just way too convenient not to sandbox it, and some tasks right away depend on not being sandboxed. For anything other than writing code directly in a fully contained g
12.
by ghrl 3mo ago
I would assume that cost to be minimal, considering their PR never got merged. And if it were me I would consider that well worth the entertainment.
13.
by ghrl 3mo ago
Or by using a proxy, yeah. Personally I would still prefer a multi provider harness over CC when using it with another provider, if alone for the visible reasoning, model switcher, cost estimation and so on. So far I've only preferred
14.
by ghrl 3mo ago
In many cases they're amazing, too. And the visible reasoning and the pricing are amazing too.
15.
by ghrl 3mo ago
Yeah, I used that too on my last Mac. But the page explicitly states the benefits of this approach (preventing it from launching all together without doing anything vs listening for the launch and killing it). It also does not use a menu ba
16.
by ghrl 3mo ago
Yes, certainly. I've heard of people that let an agent run on one machine, point a USB Camera at the target and give the agent ssh access and something like imgsnap (cli webcam command) and then let it run autonomously. The agent can t
17.
by ghrl 4mo ago
I would say if it's only used as the build and publishing device and development happens elsewhere, this would work without problems. 8Gb for building the iOS app and testing on a real device or even an emulator would likely work. Appl
18.
by ghrl 4mo ago
I was thinking all that too and considering commenting about being sick of those credit card size claims, but after seeing the footage I am genuinely impressed. Great work there.
19.
by ghrl 4mo ago
Wow, very interesting example.
20.
by ghrl 4mo ago
I believe you can get close to that with various Instagram mods for Android. They have advanced features like only show posts from followers and stuff like that. I switched to one of them after realizing the 5 second ad breaks I got were on
21.
by ghrl 4mo ago
I would disagree. Having all the messages locally and sending them with the request means you can switch inference providers or even models mid-conversation. It also means that the provider doesn't store the entire context, which often
22.
by ghrl 4mo ago
I am mostly using OpenCode and barely ever see a permission prompt. While they do enforce it for outside workspace read/write, with the bash tool the agent can just bypass that. I'm not quite sure why it is that way, and it certai
23.
by ghrl 4mo ago
I don't quite understand the benefit of the setup. If there are legacy IoT devices that need unique named 2.4G network, just broadcast another SSID for them. So each router broadcasts main 5G (common name, fast roam etc), main 2.4G (sa
24.
by ghrl 4mo ago
I would say that is highly unlikely if by SOTA models you are not just referring to coding benchmarks but more general purpose ability and domain-specific knowledge. For example Kimi 2.6, which is comparable to Opus 4.6, is roughly 500+GB l
25.
by ghrl 4mo ago
Something MacOS and Windows support natively would be a good start, it could grow from there.
26.
by ghrl 6mo ago
Let's hope they left the having-fun part for a human to do.
27.
by ghrl 6mo ago
Well, while Tauri is certainly nice, it's not quite what I imagine when I hear "native".
28.
by ghrl 6mo ago
I'm using that as well but had issues with tunneling where it creates the tunnel in the background and terminates and so you might not know the random port it assigned or I couldn't figure out how to un-tunnel it and tunnel again
29.
by ghrl 6mo ago
There used to be multiple tools like this from different websites, but they were all bought by Calligraphr to redirect to them instead, giving them an effective monopoly and letting them charge subscription fees for generating fonts over th
30.
by ghrl 6mo ago
I see a single commit adding nearly 2M lines of code with the readme claiming to be a full production-ready project. Am I missing something here? Also, since it mentions full WASM support, a web demo would be nice.

More ›