Deian Stefan
1.1K posts

Deian Stefan
@deiandelmars
Co-founder @cubistdev and Faculty at UCSD focusing on Security and Programming Languages

A novel design that is built to scale: • A Trusted Execution Environment (TEE) powered by @cubistdev, settled with onchain proof. • No smart contracts. No third-party messaging. Protocol agnostic. • Independent from third party services such as Axelar or LayerZero. • Any source, any destination. We rebuilt the rails.

Skate 1-Click is live on app.skatechain.org The best of DeFi, to you. One click. Done.

Squid Intents is live Our fully independent routing layer Cross-chain settlement that makes complexity invisible ✨ Here's how:


We’re excited to announce that @squidrouter's Cross-Chain Order Routing and Auction Layer (CORAL) v2 is using CubeSigner and Cubist Confidential Cloud Functions to enable sub-second cross-chain swaps with competitive slippage. “By leveraging Cubist Confidential Cloud Functions, we've replaced on-chain smart contracts with private, off-chain logic, enabling sub-second swaps across EVM, non-EVM, and even non-smart contract chains like Bitcoin, all with higher reliability and better pricing.” – Squid Co-Founder, @ecdsafu A unique property of CORAL v2’s architecture is that its implementation of custom transaction signing policies via Cubist Confidential Cloud Functions eliminates the need for quote expiry, a primary driver of transaction failures and price slippage in cross-chain swaps. Cubist Confidential Cloud Functions is the first programmable compute technology of its kind to reach General Availability, powering private smart contracts in cross-chain swaps and other large-scale production systems across a variety of Web3 use cases.

MetaMask is now faster than the rest. Your swaps and bridge transactions are much more competitive, because your quotes are now: 🏎️ Delivered 2x faster 🌉 Sourced across EIGHTEEN providers ⛽ No longer requiring native gas tokens for eight different networks





Lombard has acquired BTC.b—the leading Bitcoin asset on Avalanche with ~$538M in circulation. BTC.b will operate exclusively on Lombard's protocol infrastructure as a neutral public good, and introduce a permissionless, decentralized, next-generation BTC asset to the masses.



🇯🇵EXCLUSIVE: Japan's $2T Payment Provider TIS Rolls Out Multi-Token Platform With Avalanche @sndr_krisztian reports. coindesk.com/business/2025/…



I usually try not to criticize other industry players, but I still do it once in a while. 😂 This update from Safe is not that great. It uses vague language to brush over the issues. I have more questions than answers after reading it. 1. What does "compromising a Safe {Wallet} developer machine" mean? How did they hack this particular machine? Was it social engineering, a virus, etc? 2. how did a developer machine have access to "an account operated by Bybit"? Some code got deployed from this developer machine straight to prod? 3. How did they fool the Ledger verification step at multiple signers? Was it blind signing? or did the signers not verify properly? 4. So, was $1.4 billion the largest address managed using Safe? How come they didn't target others? 5. What lessons can other "self-custody, multi-sig" wallet providers and users learn from this? 🤷♂️

Policy-protected keys could have prevented the Bybit hack. Crucially, enforcing policies in the wallet backend saves users from malicious UIs. Check out example policies below (see our blog for more detail): cubist.dev/blog/understan…). Transaction policies that would prevent the attack with essentially no inconvenience in daily ops: ✅ Require 3/3 YubiKey approvals. ✅ Allow transfers only to known warm wallets. ✅ Limit transfer amounts within a set timeframe. We’re big fans of YubiKeys for phishing protection because they ensure that approvals sent to fake UIs fail. Even if signers are tricked into approving a malicious transaction, strict transaction policies enforced on the backend (not client machines!) will block malicious transfers. Governance policies that control how and when it’s possible to change the transaction policy on a wallet: 🔐 Require 5/7 YubiKey approvals for policy changes. ⏳ Enforce a 7-day waiting period, allowing time to detect and cancel attacks. A best practice is for governance changes to require more and different approvers than for transaction signing. By fully separating transaction approvals and governance approvals, attackers can’t reuse transaction approvals to approve policy changes—making exploits a lot harder. And, of course, following best practices when it comes to shipping security-critical software. Compromising a single developer's machine should not subvert the security of your whole system.



Bybit Hack Forensics Report As promised, here are the preliminary reports of the hack conducted by @sygnia_labs and @Verichains Screenshotted the conclusion and here is the link to the full report: docsend.com/view/s/rmdi832…





