Azax

🧵 THREAD TECHNIQUE: Keeta Network ($KTA) - Hybrid DAG blockchain or disguised database? 1/ After 48h of deep analysis of @KeetaNetwork testnet, here are my technical findings on this "blockchain" architecture claiming 10M TPS. I tested everything: double-spending, data corruption, wallet creation. Spoiler: it's more complex than it looks 🔍 2/ First, the context: Keeta Network presents itself as a "Hybrid DAG" with Delegated Proof of Stake (dPoS). Each account has ITS OWN blockchain, creating a virtual directed acyclic graph. Mainnet planned for summer 2025. Architecture: Account A: Block1→Block2→Block3 Account B: BlockX→BlockY Account C: BlockZ 3/ First test: Wallet creation and transactions ✅ Generated wallet: keeta_aab6gesqzex3mp7to33zknlqrnxg3n76qedude2yd6xfvravdfqdpmc22x5jvkq Received 10 tokens from faucet Successfully sent to other addresses Everything works, UX is smooth 4/ Double-spending test 💰 I tried sending the same tokens 2x simultaneously: Promise.all([ send(address1, 1 token), send(address2, 1 token) ]) Result: 1 passes, 1 fails. Basic but functional protection. 5/ Corruption test: Attempted to modify history 🔨 Result: IMPOSSIBLE. Old blocks remain intact. I even tried to: Modify hashes Inject fake blocks Manipulate timestamps Everything rejected. Immutability holds. 6/ Consensus analysis: 4 representatives, 4 votes per staple ✅ BUT... all 4 votes have the SAME signature (!!) Unique signatures: 1/4 Unique issuers: 1/4 Normal for a team-managed testnet, but raises questions for mainnet. 7/ "Client-Directed" Architecture - The most controversial point: Client directs its own validation No global mempool "Presumed not to be conflicting" I tried to exploit this. No success yet, but it's a potential attack surface. 8/ Cryptographic tests ✅ SHA256 hash: 64 characters ED25519 + ECDSA signatures Vote staples with consensus Unable to forge signatures Crypto is solid. 9/ Testnet performance: Transactions: ~1-2 seconds Finality: Near-instantaneous 10M TPS? Not tested (limited testnet) Whitepaper admits it's with 1 local validator. Reality check needed on mainnet. 10/ Fascinating technical points discovered: ASN.1 DER encoding (extensible) Native multi-token support (I created my token!) On-chain permissions (tested SEND_ON_BEHALF) Votes encoded in X.509 (!!) Functional storage accounts 11/ TECHNICAL VERDICT: ⚖️ ✅ IT IS a real DLT system (DAG type): Unbreakable cryptography Functional consensus Proven immutability Impossible to corrupt (in my tests) ❌ BUT: Centralized testnet (normal) Questionable "client-directed" architecture True decentralization to be proven on mainnet 12/ My take: Keeta Network is technically solid. The testnet works, impossible to break despite my attempts. BUT everything will depend on mainnet: how many independent validators? Open or permissioned? 13/ For devs: Excellent TypeScript SDK Generous faucet for testing Decent documentation If you're looking for a high-performance alternative to Ethereum, keep an eye on mainnet this summer 14/ Questions for @KeetaNetwork team: How to become a validator on mainnet? Real decentralization planned? Will "client-directed" remain? Post-mainnet roadmap? Like and retweet to support our work. (Cryptovest.com reconstruction in progress) #Blockchain #Web3 #DAG #KeetaNetwork #CryptoForensics #TestnetAnalysis