gandlaf21

Bitcoiner, Nostrer, Developer.

Public Key

npub1cj6ndx5akfazux7f0vjl4fyx9k0ulf682p437fe03a9ndwqjm0tqj886t6

NIP-05 Address

www.gandlaf.com

Profile Code

nprofile1qqsvfdfkn2wmy73wr0yhkf065jrzm8705ar4q6clyuhc7jekhqfdh4spzemhxue69uhhyetvv9ujuurjd9kkzmpwdejhgqglwaehxw309ahx7um5wghx26tww4hxg7nhv9h856t89eehqctrv5q3camnwvaz7tmwdaehgu3dxqcju7tpdd5ksmmwdejjucm0d5q3gamnwvaz7tmjv4kxz7fwv3sk6atn9e5k7qgqqythwumn8ghj7un9d3shjtnwdaehgunfvd5zuer9nehn3d

Publishing to

relay.primal.net

nostr.einundzwanzig.space

nostr-01.yakihonne.com

relay.damus.io

relay.nostrich.de

Author Public Key

npub1cj6ndx5akfazux7f0vjl4fyx9k0ulf682p437fe03a9ndwqjm0tqj886t6

Show more details

Last Notes

2026-09-08 10:22:16 CEST

- reply

Right! Thanks for the explanation. 🙏
because in this case, it relied on the cache, revalidation would expose the bad egg.
Agree, validation could be tricked through a bug, but it could still easily be traced if inputs and outputs don't add up.
A truly hidden inflation bug is only possible if amounts are hidden. would you agree?
(again, not saying it happend or it would happen, just talking about the possibility)

2026-09-08 00:53:44 CEST

- reply

Are you sure? Could they have crafted a tx that doesn't peg out, just inflate and then slowly drain, or even just use it to transact inside of liquid? i don't have enough technical knowhow about liquid to know if this is possible, but if it's not possible, can you explain why?
I mean on bitcoin inflation would be very easy to spot, even if it's just 1 satoshi, since amounts are not obfuscated. there is no hidden way for inflation to happen on bitcoin

2026-09-07 14:37:28 CEST

- reply

Yea, wasn't aware how liquid works. It makes sense

2026-09-06 23:11:27 CEST

- reply

tf does that even mean: they withdrew it with that PAK key, but the key wasn't compromised??
Makes no sense

2026-09-06 23:08:31 CEST

- reply

W China, i guess

2026-09-06 23:03:56 CEST

- reply

monero, zcash (sometimes?), and liquid rely on cryptograpic proofs that hide the amounts to basically say "no more coins were created here than put in". But since no one can verify the amounts itself, you rely on the cryptographic soundness of those proofs. if they have a bug that lets coins be created arbitrarily and no one knows where they come from, we call it a hidden inflation bug.
As opposed to o visible inflation bug, where you would see where coins were created illegitimately.
On bitcoin that would mean outputs are larger than inputs => visible

2026-09-04 18:49:50 CEST

- reply

based. And congrats on neither forking nor rage quitting in the third cycle

2026-09-01 07:34:17 CEST

@nprofile…78ud , are you talking about the wallets hosted by primal?
#nevent1q…mzlc

2026-08-31 13:55:49 CEST

- reply

I don't fully believe this, but as a steelman:
they first fully exhaust the "by the book" options, and once they got banned they can go to the "gray area" options.
Otherwise, if they go straight to tor, coinjoins, etc, it makes it easier for their adversaries to label them as shady

2026-08-30 22:02:25 CEST

- reply

virus gotta eat too!

2026-08-30 21:59:51 CEST

- reply

I think this is maybe too technical, and maybe also overkill for a simple thing like that, but i don't trust iot devices so here is the sovereign way to set it up:
you need: 1 cam, 1 wifi router, 1 dual nic mini PC
get any iot cam, like Tapo. set them up normally, update firmware etc.
then, put the cams on the wifi router network. this router does not need internet access.
connect that router to the mini pc, where you install frigate to view and manage cam.
connect second network interface of minipc to modem.
If you don't understand it, copy paste this into AI, and it will explain in more details why this is the better sovereign solution.
otherwise, you will rely on camera manufacturer cloud infra, and they will collect your video feed

2026-08-29 12:23:19 CEST

- reply

hate only brings more hate.
Hateful people live miserable lives

2026-08-28 18:15:53 CEST

- reply

whatever makes you happy bro

2026-08-27 18:18:52 CEST

- reply

ew twitter mentioned

2026-08-27 17:37:53 CEST

- reply

woah!

2026-08-25 21:06:07 CEST

If you haven't yet, you should probably update your mac
https://nvd.nist.gov/vuln/detail/CVE-2026-65400
https://npub1cj6ndx5akfazux7f0vjl4fyx9k0ulf682p437fe03a9ndwqjm0tqj886t6.blossom.band/bcf76f5e2469e3e8a18cbee5a98813e039812533096e9fbae6584c9d2cdf3a25.jpg https://npub1cj6ndx5akfazux7f0vjl4fyx9k0ulf682p437fe03a9ndwqjm0tqj886t6.blossom.band/dfcd614730795e5a46e8bc5c2b1de32ee141e2742d9c40353d9a4f70400846e9.jpg

2026-08-25 10:42:08 CEST

- reply

nostr/divine was in the local newspaper yesterday

2026-08-25 10:16:13 CEST

- reply

there were signs

2026-08-25 10:11:25 CEST

you can brute force windows logins at 20 GH/s on a midrange laptop GPU.
That's 20 Billion a second. The login hashes can easily be extracted with physical device access.
Windows is trash tier. protect yourself
#nevent1q…rwzf

2026-08-24 23:04:55 CEST

I haven't touched Microsoft windows in a decade.
I always though they would probably have pretty bad security.
But today, I was blown away by the fact they use FUCKING MD4 TO HASH THEIR PASSWORDS💀
MD4!!

2026-08-18 03:48:34 CEST

- reply

Security issues always exist, because security is a bottomless barrel.
If you read the post, the issues they found are "not that bad". Requires a sophisticated, targeted attack to have a chance for exploitation.
if i had to guess, the attack would be to create custom firmware, get fool to flash their device with it and extract seed via malicious signatures (dark skippy)
No where near the coldtard entropy bug. Still bad tho, but there is a whole class of devices that have no protection against this class of attack

2026-08-17 15:53:15 CEST

- reply

loop audiotrack to keep the browser tab alive 💀

2026-08-17 06:03:13 CEST

- reply

they're e2e encrypted, so nothing to see there
nip-4 metadata leaks are a problem even without backdoor, but yeah nip-17 should be mostly fine

2026-08-17 02:07:43 CEST

- reply

relays don't hold sensitive data, right? so what do you mean by backdoor

2026-08-14 16:04:34 CEST

- reply

INVEST

2026-08-14 14:21:03 CEST

- reply

2026-08-14 14:20:22 CEST

- reply

i asked the AI about this, and the AI said this is FUD

2026-08-14 03:57:08 CEST

- reply

Damn, I hate tabsoco. makes everything taste the same. But i guess if you love Tabasco taste, that's a good thing 😁

2026-08-14 02:09:37 CEST

- reply

can humans produce optimal code?

2026-08-13 02:10:17 CEST

- reply

Woah, awesome! Readme seems kinda cluttered with slop, but maybe that's just me

2026-08-12 04:38:06 CEST

- reply

not the first time it happens , and it will happen again and again

2026-08-12 02:04:09 CEST

- reply

random.gandlaf.com/chapters/coin-tosses

2026-08-10 04:26:50 CEST

- reply

"the currency has to be community owned"
what does that mean? Sounds like small scale fiat

2026-08-10 04:23:52 CEST

- reply

here are your answers:
- yes
- yes
- what?
- huh?!
you have clearly no clue how any of this works, which is fine, but maybe you should learn some more about it before you form strong opinions.
I'm not disagreeing with you, your text just doesn't make any sense

2026-08-09 16:05:59 CEST

- reply

seedy in context of redlight district sounds hella weird 😂

2026-08-09 09:33:13 CEST

it's rage-quit season

2026-08-08 19:35:52 CEST

- reply

well, if we're gonna talk about this bug,
it was introduced by doc-hex aka switck aka peter gray, who wrote both the lib and the integration, though he wrote the lib under a pseudonym for whatever shady reason, and did a whole thatre talking to himself in the prs when doing "reviews"..
they were warned by multiple people (incl. jamesob) that this setup looks buggy, if not straight up dangerous, but they ignored it, waved it off, either due to hubris or malice.
It sounds to me at least 50% likely it was an attack.
but either way, you say the hacker found it with AI, but the devs couldn't? what's the logic here?

2026-08-08 19:01:19 CEST

- reply

it's bitcoin christus vision

2026-08-08 18:57:34 CEST

- reply

I don't think it's theater, it's just a different perspective on the matter.
You (vitor) clearly have the blue team goggles on, and fuzzstone is looking at it from a red team perspective.
the scenario you described is a very distinct attack, not a random bug that happens to be exploitable.
And what fuzz said holds true. if you cannot nudge and hint the AI, it most likely will be useless. The tool is only as useful as the one who wields it.
But of course i understand your perspective as a maintainer, you want to protect against that specific attack, and llms might not be particularly good qt that one yet

2026-08-08 17:49:42 CEST

- reply

what did you try? What was the strategy?

2026-08-08 16:24:31 CEST

- reply

trained on Twitter threads for sure

2026-08-08 13:00:42 CEST

- reply

so close! you were really fast too

2026-08-08 13:00:22 CEST

- reply

nice!

2026-08-08 12:59:30 CEST

solved by @nprofile…980n 🍻
#nevent1q…enfr

2026-08-08 12:58:52 CEST

- reply

solved by @nprofile…980n . congrats!

2026-08-08 11:33:55 CEST

Crack this NSEC first and I will zap you 10K sats!
To win, find the flag after logging in with the compromised nsec.
Then, comment the flag under THIS post, with your real npub, so I can zap you.
#nevent1q…02zr

2026-08-08 05:02:46 CEST

https://xcancel.com/i/article/2085865561137831938
https://npub1cj6ndx5akfazux7f0vjl4fyx9k0ulf682p437fe03a9ndwqjm0tqj886t6.blossom.band/df6861e73b412a54009bbb74a95c7e7fd263e4f7bf3abc965399142b7331dbe9.jpg

2026-08-08 02:18:16 CEST

anyone already tracking the theft?
Allegedly, this is the attackers address
19jSfadPiKa4aCefj4F7uEYbhMEt2L8SCH
according to this post
https://xcancel.com/Gioluz_/status/2085805228239081616#m
#nevent1q…fpnu

2026-08-07 18:08:56 CEST

- reply

https://npub1cj6ndx5akfazux7f0vjl4fyx9k0ulf682p437fe03a9ndwqjm0tqj886t6.blossom.band/52c155eaa7dee5aa32235a5aaed60ba7ebd1735e3f35ae6101953c5e5858b960.jpg

2026-08-07 16:56:57 CEST

when SBF blew up FTX, he didn't hide

2026-08-07 08:08:30 CEST

- reply

everything is temporary ☀

2026-08-07 01:59:06 CEST

- reply

device id is bad, this info is findable by a motivated attacker, especially if the device id has a pattern (i think in cc case it was 16bit id, but could possibly be reduced to lower entropy contribution based on pattern, or even zero of the attacker knew the ids)
timestamp is bad, because the possible values are a tiny set, and they are highly predictable
keypresses are ok, but they were not part of the standard setup, only if you rolled dice
good one was supposed to be seeded with environmental noise captured by the secure element, allegedly

2026-08-07 01:20:48 CEST

- reply

that assumption was already established by black hats as soon as cc exploit started happening

2026-08-06 16:36:30 CEST

- reply

it's groundhog day. again.

2026-08-06 12:00:21 CEST

- reply

i always wanted to know what it feels like to be the kning of zuxbux
... Better start grinding!

2026-08-06 08:25:16 CEST

- reply

software can't generate entropy. but hardware can capture it, and hand it off to seed the software. The weak link in this pipeline is the human implementing this correctly.
so, better off to simplify thr pipeline and throw some coins
https://random.gandlaf.com/chapters/coin-tosses

2026-08-05 08:47:28 CEST

- reply

anyone able to go through the diff 4.0.0..4.0.1 and see what bug they fixed?

2026-08-05 05:21:49 CEST

- reply

the very first one has the 31 min, instead of a date.
yeah i saw that you mentioned it that's why i was looking if i could tell the shop, and that was the thing that stood out :)

2026-08-05 05:06:27 CEST

- reply

sorry, meant to say signature nonces, not the other way around hahah :) thx for the zap!

2026-08-05 04:26:02 CEST

- reply

not a twitter user myself, but timestamps look wrong (maybe due to PS)?

2026-08-05 03:06:54 CEST

- reply

thanks!

2026-08-05 02:35:41 CEST

- reply

shit entropy is a vulnerability, but you usually can't just do a search for "keys with bad entropy".
In the CC case, being aware of the bug was a pre requisite, because the attacker needs to be able to identify the target search space, which is only possible if he knows the (pseudo random) key generation process and the potential inputs (2^40)
So the searchspace is a window of 2^40 , but the window is somewhere in 2^128, and without knowing the bug, you can't know the location of the window, so the search space remains 2^128
there is exceptions, where bad entropy itself can reveal a bug, but i think it's mostly related to bad entropy in nonce signatures

2026-08-05 02:17:55 CEST

- reply

I think even the seedsigner community was bummed out by this news

2026-08-04 10:04:13 CEST

new hot shit from krux dev:
https://github.com/odudex/Kern

2026-08-04 08:08:45 CEST

Time to flip some coins!
https://random.gandlaf.com/chapters/coin-tosses

2026-08-04 06:00:55 CEST

- reply

if you input the coinflip bits into the browser, it may cache/save the input value. depends on the browser probably

2026-08-04 04:13:23 CEST

- reply

btw in the lookup table, it says bits 1-6, but should be bits 3-6.

2026-08-04 03:37:25 CEST

- reply

neat way to get around decimal transform!

2026-08-04 02:51:11 CEST

- reply

hahah nice

2026-08-04 02:09:53 CEST

- reply

wrong lesson learned

2026-08-04 00:43:29 CEST

- reply

Indeed, every signing device or wallet can do these things (checksum, derive address)
and if we don't trust a single one to derive the correct addresses, we try multiple. Of course they all have to be airgapped.
Your tool i assume could be one of those, to cross check.
Some things to watch our for:
Browser fields may cache unwanted inputs
The delivery html should leave the code uncompressed for auditability
browsers can behave differently
for me personally i would feel uneasy using the web browser for this... even on an offline device.
maybe an unfounded fear, just because browsers are so huge and clunky

2026-08-03 17:24:56 CEST

- reply

why would you need to implement anything? all you need is a coin and the wordlist

2026-08-03 15:40:15 CEST

- reply

now, do you trust the coin? Maybe your coin has a bias to land on head, because of uneven weight distribution. A bias can have a negative effect on the entropy of the bits produced. studies say average coin bias is 51/49, which is low enough to not be an issue.
If you still don't trust the coin, you can use von neumann normalization.
For each throw, you do 2 throws instead:
if
heads, heads => rethrow
tails, tails => rethrow
heads,tails => heads
tails, head => tails
this will require more than double the throws, but will give you unbiased entropy

2026-08-03 15:28:16 CEST

- reply

you can create a random seed phrase (mostly) without software.
All you need is the wordlist (with indexes) and a coin.
flip the coin 11 times. this will give you 11 bits. transform the bits into decimal number index and lookup the word in the wordlist,
boom, first word down
repeat 11 times.
for the 12th word, flip 7 times. The last 4 bits of the last word are the checksum.
Some airgapped signers like jade will provide simplified input, so you don't have to try to find a valid last word 16 times. This is the only step that requires software because it relies on sha256, but this step does nothing to the randomness, so there is no implementation you have to "trust"

2026-08-03 11:55:52 CEST

- reply

why must you hurt me like this

2026-08-03 11:38:27 CEST

- reply

this depends on the attack,
But if the attacker knows the seed phrase, the passphrase needs to have ~70bits of entropy.
Per truly random alphanumeric character, you get about 6bit per character. so you need 12 minimum
If the passphrase is not truly random, the bits you get per character sink drastically (eg. it includes well known sequences or numbers)

2026-08-03 09:33:24 CEST

I forgot who said it, but i thought it was an interesting take:
The coldcard attacker may have not swept sub .15 coin wallets because he labeled them "potential decoy/canary" . Therefore he may have grinded some passphrases first, before sweeping the canary... in hope of finding the real stash.

2026-08-03 02:42:06 CEST

- reply

what's the benefit of using slipstream?
I thought RBF race only applies to multisig, since you would expose descriptors?
and doesn't using slipstream effectively slow down the transaction, since you have to wait for Mara to mine a block?

2026-08-03 01:24:09 CEST

- reply

i think i understand now what he meant: "If that security assumption turns out to be inadequate, a lot of software will be broken".
which is true.
But yeah it's fud, imo

2026-08-03 01:01:51 CEST

- reply

good enough for what? 50 dice rolls is 129ish bits of entropy, 2^129 equally possible outcomes.

2026-08-03 00:54:45 CEST

#nevent1q…qsat

2026-08-02 10:53:10 CEST

- reply

pop-shove-it-in-ur-face

2026-08-02 04:32:11 CEST

https://xcancel.com/KLoaec/status/2083631322539470848

2026-08-01 07:40:43 CEST

- reply

there are official announcements... it's not FUD

2026-07-31 19:03:23 CEST

- reply

https://xcancel.com/glxyresearch/status/2083181683067506899#r

2026-07-31 19:03:02 CEST

- reply

https://npub1cj6ndx5akfazux7f0vjl4fyx9k0ulf682p437fe03a9ndwqjm0tqj886t6.blossom.band/da10204e838d8028352c6994c82713f7507089ac019e6b585c995a3aabbae8b8.jpg

2026-07-31 14:41:54 CEST

- reply

yeah that's where the funds were moved:
in this tx: 0c6bf853a645b699a3b2cd6d8e3c44cf1a02a16f538df08212a44753f75d9d01
it's the same heist

2026-07-31 08:20:14 CEST

- reply

isn't it the same address? it used to have ~600 bitcoin, but some of the funds have moved been moved

2026-07-26 08:01:47 CEST

- reply

1 issue = 1 sat

2026-07-26 05:42:07 CEST

- reply

tastes like FMJ? 😁

2026-07-21 12:41:58 CEST

- reply

your zap address ia broken

2026-07-20 01:53:43 CEST

- reply

the good guys won

2026-07-09 06:08:46 CEST

- reply

Really like this style of video, well done!

2026-06-29 05:25:43 CEST

- reply

remember Epstein?

2026-06-25 05:13:33 CEST

- reply

and Paul Sports coin too!

2026-06-24 09:24:48 CEST

a wise man once said:
"Money ain't nothing but garbage"

2026-06-18 04:32:16 CEST

- reply

100 sats per outgoing mail seems like enough to prevent spamming, or no?

2026-06-15 14:43:21 CEST

- reply

just one more

2026-06-14 03:49:14 CEST

- reply

yes and no, there is no way to tell who is a user of a mint, so in theory it could be anyone.
but practically speaking, if you use it (just you) to make LN payments from your node, it doesn't change the privacy dynamics

2026-06-05 08:52:30 CEST

- reply

these people have still not learned lesson one of hodling:
stay humble