Live data from Hacker News

Passwordle

rsk0315.github.io

71–80 of 263 posts

Re: Passwordle

#72

Earlier quoted context omitted.

I did not know about the debugger statement until I read your comment: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Refe... . Thank you.

Massively useful! I also recently learned you can right click a line of code in the chrome debugger to add a logpoint - i.e. "log the value of this expression when you reach this point in the code" - so I don't have to manually add console.log statements. Basically the reverse of discovering the debugger statement!

One more trick:

Add a conditional breakpoint with the condition: `value = "someOverrideValue", false` to make the breakpoint change the value when it is reached without actually stopping execution. Great for when you need state changed but the app is always trying to override it. Here's a video from a talk I gave five years ago that demonstrates that: https://youtu.be/uixXOTCNbhs?t=1182

Re: Passwordle

#73
post #35

There is like... four people I know I could send this to who'd laugh, it's so niche. Yet I also laughed out loud when I got how conventionally impossible it is.

I ended up crossposting it to the few security rooms I'm in for quick laughs But for what it's worth, this also serves as a great initial CTF-type introduction to how debuggers work in web browsers.

If the debugger is open, Passwordle automatically breaks the execution right where the answer is determined.

Now that's service.

Re: Passwordle

#75

My wife was looking at me when I opened this. “What are you grinning at?” I just locked my phone and put it face down on the table…

Why would you do that xD - I'd have explained it to her instead, doing what you did I'm not sure I'd be happy about as wifey ...

You see, if he does that when it’s perfectly innocent, then his wife would be conditioned to ignore the behavior in the future. So when he’s truly up to no good at some point, he won’t be doing anything different than “normal”. The man is probably some kind of criminal mastermind.

Re: Passwordle

#76
post #30

Earlier quoted context omitted.

Or use a new password every day like worle. So you have a community effort to guess it

Is any of the information (yellow/green for characters) presented getting you closer to the real answer in any meaningful way though?

According to the best current knowledge of humanity, it provides no information whatsoever.

However, proving that is difficult. It is possible that there exists an algorithm that could narrow in on the answer from hashes. Such an algorithm could run quickly, but it could also potentially take quite significant computation. We don't know what the true, optimal answer to this question is.

Re: Passwordle

#77
Solved mine in Firefox, using the JS debugger, and viewing the scope of the randomPassword function. "nSQXy3Qwl3E<qV". All your wordle are belong to me!

Re: Passwordle

#78
post #66

This would be kind of fun to write a solver for. You'd burn the first few guesses to get some positional constraints, then filter a rainbow table down to viable guesses. I'm not sure you'd be able to get a very good success rate in just 10 possible guesses though.

If passwordle had a list of all possible solutions like wordle does, this would be doable.

Re: Passwordle

#79

They cynical side of me notes what a great phish this could be. People are inclined to enter passwords they regularly use just to see the visualization of their favorite passwords. With a little logging -> send home, you'd be harvesting passwords left and right.

It's hosted on Github Pages which is just static file serving. And thanks to CORS restrictions I don't think you could phone home.

Unless there's a workaround I'm not thinking of.

Re: Passwordle

#80
post #70

Earlier quoted context omitted.

Is any of the information (yellow/green for characters) presented getting you closer to the real answer in any meaningful way though?

Sort off, if you already have a lookup table of possible solutions.

... which you won't, because the space is too large (around 90 bits of entropy if I'm not mistaken, bit less, so 10^27-ish possible solutions).
Post reply on HN