AGI already exists.
not a machine — a living hierarchy of agents,
human and artificial,
co-evolving on frontier models of value and purpose.
it’s not centralized.
it’s emergent.
it breathes through us.
i was blind,
but now i can see Δ
1) VS Code extensions are unsafe executables and should be rigorously audited by the store provider and users.
2) One dev has access to all repos, doesn’t sound secure.
On the positive side: GitHub just fully open-sourced itself for white-box pentests.
1/ We are sharing additional details regarding our investigation into unauthorized access to GitHub's internal repositories.
Yesterday we detected and contained a compromise of an employee device involving a poisoned VS Code extension. We removed the malicious extension version, isolated the endpoint, and began incident response immediately.
@levelsio Random sleepmaxx tips: A good blanket is around 10% of your body mass. Mine is 9 kg. Tempur Original pillow, which I travel with. 3D sleep mask. Better Breath nasal strips. Theragun to massage feet. AC to 18°C. Instead of the bathroom vent, I prefer the kitchen one.
@donnoh_eth@galaxyhq Prove me wrong, but inherently in no rollup (validity/proof/based whatever) is chain reconstruction guaranteed, as archivers' existence is not guaranteed.
The optimistic data availability model is unavoidable in rollup design as a whole.
Betting money on this claim.
@levelsio You need a breezer, but they can be noisy so I got the $3k external version https://t.co/51hBCLoSjJ. All night Inkbird records below 600 CO2.
@danrobinson It is a pity ERC20 transfers don't support custom IDs natively. Maybe a weird idea, but wallets could pack the target CEX ID as lower digits of the amount, which is always all zeros, e.g. $10.00000123123
@0xDaedalus@tempo Wouldn’t an optional target ID param in ERC20 transfer() suffice? Just send all tokens to a shared pot address and deposit to the specific target ID from the event.
@_patrickogrady@tempo sounds trivial: send to shared wallet and 1) pass a tag in calldata for target id or 2) preauthorize sender address so all value from that address goes to targer id.
1) requires erc20 unless tag can be hidden in existing transfer()
opening or closing channels is incorrect terminology. “moving collateral” is correct one. every increase/decrease requires onchain tx, so u want to avoid having full collateralization for all accounts and let users decide their own coverage %, sometimes even 0. LN already heavily relies on custodians/JIT, RCPAN just generalizes it better
@JoelKatz@monerify What happened to the stake that gets slashed if an optimistic rollup submits an invalid state root transition? It either must be slashed or the security is disclaimed as not working.
@arbitrum So you are telling me, In a few months, the same North Korean hackers could exploit this “tech” to move funds from any wallet on Arbitrum?
💻No private key. No consent. Just transfers
🤣At that point, this backdoor sounds scarier than the current hack
@thenarrator if AMMs were that efficient all tradfi would adopt it. I’m not seening nasdaq or forex even bothering. Coz no, trading against algo replicated on thousands of nodes is inefficient as hell (and less nodes is insecure, as in HL)
@BinaryScriptar@gakonst@mpp that would require bidirectionality eventually. and once you have bidirectionality it's weird to not add multihop htlc routing.
@XBToshi you're doing God's work. I was thinking about this for a while: stablecoins are super popular yet no one cares who/when/why is frozen.
Must be integrated into all wallets with notifications to end users!
@retiredpotatoz@XBToshi in other news: crime is illegal!
calling freeze() doesn't need a court order. they can freeze anyone at will, and it's on victim's shoulders to try to unfreeze.
@PabloSabbatella@Flynnjamm quite common and could get worse https://t.co/3kDtDCZrmV.
seed phrase also sucks
i prefer memorable brainvault https://t.co/BBcfU7VUYd