arshxyz

> a typical client-server exchange creates a trail of user data, like the client’s IP address or TLS fingerprint. This level of visibility can be a burden.

Does Cloudflare's WAF (which relies on TLS Fingerprinting) stop working if OHTTP is enabled? If not, does this imply the client metadata is read and processed by Cloudflare but not passed on to the application server?

<50MB .dmg and no electron? I'm sold

Who needs Sonnet 12817 when 5 is maintaining is already maintaining Omarchy? And it's only 5% JS!

Cloudflare WARP suffered a similar downgrade recently when it went from a nifty menubar app on macOS to a vibecoded window built with flutter

For some reason slop UI labels always refer to the user in second person. Solid tell when something says "Needs You" instead of "Requires Attention"

This used to happen a lot with GPT 5.4. It would start outputting entire sentences in Korean for no apparent reason.

Cost per Token is not representative of the actual cost incurred because a cheaper model can waste a lot of tokens to accomplish the same task. Also this does not consider cost for cached tokens which can often make a "cheaper" model more expensive even when total tokens spent are constant

I had Jev play a bunch of games testing:

- Various encoding schemes

- Belief tracking

- Susceptibility to prompt injection

It was surprisingly good at decoding Base64 but failed at something far simpler for humans: ROT13.

I don't understand the point of putting visible Agent Instructions on the page taking up real estate while a human has the page open in a browser. Won't an agent be able to trivially extract these instructions if they came with a display:none rule?

Amazing how many products exist purely due to Google's lack of motivation to do things right. Will be using this a lot.

Does Google Apps Script expose an API to do this though? It would be easier to download an html page from within GDocs without having to run something locally.

> Start llama-server on your model and run a prompt

If this is supposed to target closed-weight models it would be naive to assume they will work out of the box with llama

way more fun version of this went viral on twitter a while ago: https://youraislopbores.me

3 different shades of blue intersecting each other representing different models. If you want a graph to be easier to read use colours that aren't visually similar

Dreadful color-coding on the output tokens table

They seem to be using the provider's harness for each

I extracted the zip, found the electron build folder, replaced a string in the minified code to launch Chrome on macOS correctly and ran electron-packager with the target being macos.

Patching electron apps is fairly common. You can take a look at Spicetify or BetterDiscord to see the process in more detail

I was able repackage it for macOS. I can help you test it too. Shot you an email at contact@apiparrot.com

Interesting project, I've often looked for something like this but haven't found anything that does the job. I'm on a mac and can't wait to try this out. Can I ask what you're using adblock-rs for?

They fixed the issue but for a beautiful moment in time, it looked like this: https://archive.is/LPC5O

"Nah the build quality is great but I just can't handle a functioning notification system that doesn't come with spam"

That is clearly an altered image. The padding is inconsistent and the border radius on the highlighted text is wrong

Half of these are fake/satire. Look at the first one where the highlighted text colour should be white and not tinted red.

The cp3 one has inconsistent padding. The astronaut one is clearly inspect element. I could go on

It is compatible with ollama. Advanced settings -> "URL for AI API"

> HTML to awkward so you have to switch to webgl which is too low level or you might switch to svg

HTML is replaceable by WebGL which is replaceable by SVG?

> JavaScript that compiles down into typescript

JS compiles down to TS?

I am not sure if anything in this comment indicates you have any experience with developing for the web but that is not my primary issue with this comment. My gripe is claiming that things are "awkward" or can't be done without even looking around. None of the examples below use WebGL

Here's a WinXP UI clone that runs in your browser with floating, resizable windows. - https://winxp.vercel.app/

Here's one that mocks macOS - https://macos-web.app/

Google did make something like this: https://www.google.com/advanced_search

It has a lot of options, even allows you to search ONLY the Text/Title/URL and you can filter by usage right (free for personal/commercial use etc)

Interestingly it was last updated before material design was a thing. Maybe it's nostalgia but I love the design language of Google had before material injected padding everywhere.

This is likely very paranoid behaviour but I recommend downloading all your Google data from time to time: takeout.google.com

Related: https://dns.toys

The devs behind tldraw also played with this to create something cool - https://twitter.com/steveruizok/status/1727625036159234555

As you mentioned you like Alfred, why did you choose to build a standalone app as opposed to an Alfred extension? Since you use Electron: spinning up a browser every time I need a quick-launch bar seems a little overkill.

Anyway, the UI looks very neat! Starred the repo and will try playing with it.

throwaway914, would you be willing to send over your drivers license for verification to participate on HN?