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

Why did you need to break the spelling of Iraq and Iran?

I believe they're trying to highlight the character to make clear the difference in the two similar names

Assuming the codex editor is the editor for the area below the auction counter, isn't that a security vulnerability that can put the site audience at risk?

The Codex agent is only given tools to edit the single HTML file that displays on the homepage. The page is on a separate domain, so there's no cookie sharing, and the iFrame is in a sandbox. That said, the biggest risk is social engineering attacks.

What’s to stop someone rewriting the iframe wrapper to hide the real iframe and display a fake one?

They cannot edit the iFrame itself. The user is allowed to edit the contents within the iFrame.

>The GLP-1 tsunami is approaching. As demand for weight loss drugs surges, some food and drink categories are set to decline sharply.

Defeated by either noscript or mobile's double-tap to select, copy.



Will need to test later but minor comments: was expecting an intro or FAQ so I can learn what this is about before getting into the action.

Also maybe not a good idea to post user email addresses in the leaderboards.


Thank you for the feedback. Noted.

These are some great ideas, and will be added.


-Personal comment opener -Paragraph with English words that loosely fit together yet fails to convey an idea -Heavy and predictable em dash use

I'm going to miss the days where these signs were obvious.


As interesting as this sounds and not to devalue your work, but this reads more like "Tell HN" rather than "Show HN". Can you offer any visualizations to help someone understand what you're working on?


Fair point — thanks for calling that out.

I didn’t include visuals initially since this is still research code, but I added two high-level, conceptual artifacts to make the work more concrete (no implementation details):

Architecture overview: https://gist.githack.com/rahuludacity/d787343ca72be97ea1ae51... Illustrative case study (signal vs price divergence): https://gist.githack.com/rahuludacity/be9fd41193b96c4061bf00...

The goal of both is just to show the shape of the system and the kind of signal it surfaces, not to make trading claims. I posted early because I’m still deciding which graph layers/edges are actually worth keeping before locking in the visualization layer. Very open to feedback on whether these visuals make the problem clearer or if there’s a better way to “show” this kind of system.


If an LLM was intelligent, wouldn't it get bored?


Why should it?


Lots of empty space between a posts. Less detailed than a HN entry. The boring flat/opaque design style that is plaguing UI's is captured perfectly.

It's everything I don't want or need!

Also, what does the OPP acronym stand for? Opinon has 1 “p“ (/s is that you GPT5.2?)


Thank you for your response. I'm a real person, not a GPT5.2 bot. Thanks


A bit fluffy of a piece.

>Unlike trading money for food at the supermarket where you and the supermarket both know for certain you each benefit

There was a boycott movement in Canada for due to price gouging by the grocers and I've seen 100-200% price increases in certain items in the last 4 years where I'm at (while 0% on others). Someone may be benefiting more..

>You wouldn’t wager money on your ability to beat Djokovich in tennis or Lebron James in basketball, why would you wager money on your ability to beat Citadel?

I don't believe this closing statement is accurate or relevant: you are not directly up against anyone except for yourself. Your strategy and decisions will lead you to profit or peril; against, and just like everyone else.


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

Search: