https://crypto.training/ Web3 security engineering notes: Solidity, EVM internals, audits, MEV, bridges. en https://crypto.training/blog/2026-03-04-bonding-curves/ https://crypto.training/blog/2026-03-04-bonding-curves/ Wed, 04 Mar 2026 00:00:00 GMT A comprehensive deep dive into bonding curves — from the integral calculus behind pricing to 17 real protocols (Uniswap, Curve, Balancer, Bancor, Friend.tech, Intuition, pump.fun, VRGDA, and more). How each curve shapes incentives, who benefits, and what breaks. https://crypto.training/blog/2026-03-03-eip-8141/ https://crypto.training/blog/2026-03-03-eip-8141/ Tue, 03 Mar 2026 00:00:00 GMT Vitalik’s 2026 account abstraction thread reframes EIP-8141 as a minimal protocol primitive: frame transactions with sender and gas-payer authorization. Here is the short, practical breakdown. https://crypto.training/blog/2026-02-19-smart-account-modules/ https://crypto.training/blog/2026-02-19-smart-account-modules/ Thu, 19 Feb 2026 00:00:00 GMT A code-first deep dive into Safe modules, Zodiac, ERC-7579 architecture, and major module implementations (Safe7579, Kernel, Nexus, OpenZeppelin, Rhinestone Core). Focused on feature power versus security risk with concrete Solidity references and dense Mermaid diagrams. https://crypto.training/blog/2026-02-18-smart-accounts/ https://crypto.training/blog/2026-02-18-smart-accounts/ Wed, 18 Feb 2026 00:00:00 GMT A code-first, risk-first deep dive of Coinbase, MetaMask, Thirdweb, Nexus, Light, Kernel, Safe, SimpleAccount, Solady, and Trust smart accounts: signature models, module systems, upgrades, replay controls, and security tradeoffs. https://crypto.training/blog/2026-02-17-eip-4337-evolution/ https://crypto.training/blog/2026-02-17-eip-4337-evolution/ Tue, 17 Feb 2026 00:00:00 GMT A code-first evolution map of ERC-4337 EntryPoint releases: what changed in each version, why it changed, what new risks were introduced, what audits fixed, and what 'battle-tested' really means in practice. https://crypto.training/blog/2026-02-14-auditor-operating-system/ https://crypto.training/blog/2026-02-14-auditor-operating-system/ Sat, 14 Feb 2026 00:00:00 GMT Auditing is a craft with artifacts: models, maps, experiments, and writeups. This post is an 'operating system' for doing web3 security work that scales beyond vibes. https://crypto.training/blog/2026-02-13-defi-incident-patterns-oracle-rounding/ https://crypto.training/blog/2026-02-13-defi-incident-patterns-oracle-rounding/ Fri, 13 Feb 2026 00:00:00 GMT Cream, PancakeBunny, and bZx are not three unrelated stories. They are variations of one meta-pattern: untrusted external state crossing into accounting. Here's how to spot the variants. https://crypto.training/blog/2026-02-12-transaction-forensics-tx-graph/ https://crypto.training/blog/2026-02-12-transaction-forensics-tx-graph/ Thu, 12 Feb 2026 00:00:00 GMT Block explorers show what happened, not why it happened. This is a practical workflow for reconstructing intent using receipts, logs, and call graphs, with two mainnet transactions as examples. https://crypto.training/blog/2026-02-11-rounding-in-defi/ https://crypto.training/blog/2026-02-11-rounding-in-defi/ Wed, 11 Feb 2026 00:00:00 GMT Integer math is deterministic. Your rounding policy is not. This post connects fixed-point arithmetic, share accounting, and real exploit patterns where dust becomes profit. https://crypto.training/blog/2026-02-10-uniswap-v4-hooks-security/ https://crypto.training/blog/2026-02-10-uniswap-v4-hooks-security/ Tue, 10 Feb 2026 00:00:00 GMT Hooks let you extend AMMs with custom code. They also move your security boundary into a callback. This is how to design hooks that survive reentrancy, MEV, and rounding edge cases. https://crypto.training/blog/2026-02-09-auditing-workflow/ https://crypto.training/blog/2026-02-09-auditing-workflow/ Mon, 09 Feb 2026 00:00:00 GMT Good audits are structured: define invariants, map entry points, test the scary parts, and write findings that lead to fixes. https://crypto.training/blog/2026-02-08-engineering-zk-prover-pipelines/ https://crypto.training/blog/2026-02-08-engineering-zk-prover-pipelines/ Sun, 08 Feb 2026 00:00:00 GMT ZK systems fail in boring ways: nondeterminism, mismatched parameters, and opaque errors. Build pipelines that make proofs reproducible and auditable. https://crypto.training/blog/2026-02-06-audit-checklist-erc20-hooks/ https://crypto.training/blog/2026-02-06-audit-checklist-erc20-hooks/ Fri, 06 Feb 2026 00:00:00 GMT Treat token transfers as adversarial. Measure deltas, expect lies, and design accounting that survives hooky or weird tokens. https://crypto.training/blog/2026-02-05-foundry-security-testing/ https://crypto.training/blog/2026-02-05-foundry-security-testing/ Thu, 05 Feb 2026 00:00:00 GMT Foundry turns security ideas into executable checks. Fuzz what you fear, encode invariants, and keep exploit tests as regression armor. https://crypto.training/blog/2026-02-04-upgradeability-and-initializers/ https://crypto.training/blog/2026-02-04-upgradeability-and-initializers/ Wed, 04 Feb 2026 00:00:00 GMT Upgrades multiply risk. Most upgrade exploits are not fancy; they’re misconfigured initializers, storage collisions, or privileged upgrade paths. https://crypto.training/blog/2026-02-03-evm-gas-myths/ https://crypto.training/blog/2026-02-03-evm-gas-myths/ Tue, 03 Feb 2026 00:00:00 GMT Most protocol failures aren’t about stealing funds. They’re about stopping progress. This is the gas-side of that story. https://crypto.training/blog/2026-02-01-oracles-and-twap/ https://crypto.training/blog/2026-02-01-oracles-and-twap/ Sun, 01 Feb 2026 00:00:00 GMT Oracles are an interface between economics and code. If you integrate a price, you must model who can move it, how fast, and at what cost. https://crypto.training/blog/2026-01-29-ethereum-dark-forest-mev/ https://crypto.training/blog/2026-01-29-ethereum-dark-forest-mev/ Thu, 29 Jan 2026 00:00:00 GMT Assume adversaries see your transaction before it lands and can reorder around it. This is the starting point for MEV-safe protocol design. https://crypto.training/blog/2026-01-26-bridges-threat-model/ https://crypto.training/blog/2026-01-26-bridges-threat-model/ Mon, 26 Jan 2026 00:00:00 GMT Bridges fail at the boundary between chains: verification, finality, replay, and liveness. Treat bridges as protocols with their own consensus. https://crypto.training/blog/2026-01-23-reentrancy-beyond-the-basics/ https://crypto.training/blog/2026-01-23-reentrancy-beyond-the-basics/ Fri, 23 Jan 2026 00:00:00 GMT Reentrancy is not just 'send ETH then update state'. It’s any time external code can observe or mutate your state mid-invariant. https://crypto.training/blog/2026-01-20-signatures-eip712-permit/ https://crypto.training/blog/2026-01-20-signatures-eip712-permit/ Tue, 20 Jan 2026 00:00:00 GMT Signatures move authorization off-chain. The bugs are subtle: replay across domains, nonce misuse, and integrations that assume permit always succeeds. https://crypto.training/blog/2026-01-14-solidity-storage-layout/ https://crypto.training/blog/2026-01-14-solidity-storage-layout/ Wed, 14 Jan 2026 00:00:00 GMT Storage is where Solidity hides complexity. If you can reason about slots, you can reason about upgrade safety, state corruption, and exploit primitives. https://crypto.training/blog/2026-01-08-invariant-first-auditing/ https://crypto.training/blog/2026-01-08-invariant-first-auditing/ Thu, 08 Jan 2026 00:00:00 GMT Static checks and fuzzing are necessary but insufficient. Invariant-first auditing makes protocol safety properties explicit and testable across long adversarial sequences. https://crypto.training/blog/2026-01-07-metamorphic-contracts-cross-chain/ https://crypto.training/blog/2026-01-07-metamorphic-contracts-cross-chain/ Wed, 07 Jan 2026 00:00:00 GMT Address-based trust assumptions fail when runtime code can change behind the same address on chains with divergent SELFDESTRUCT semantics. https://crypto.training/blog/2026-01-06-liquidation-as-security-property/ https://crypto.training/blog/2026-01-06-liquidation-as-security-property/ Tue, 06 Jan 2026 00:00:00 GMT Liquidation logic is often treated as pure economics. In practice, it is a security boundary where liveness failures and asymmetry bugs convert stress into protocol loss. https://crypto.training/blog/2026-01-05-mellow-architecture-and-workflows/ https://crypto.training/blog/2026-01-05-mellow-architecture-and-workflows/ Mon, 05 Jan 2026 00:00:00 GMT A code-level deep dive into Mellow Flexible Vaults: core modules, queue-driven fund flows, curator execution, oracle reporting, and security boundaries. https://crypto.training/blog/2026-01-02-adversarial-token-integration/ https://crypto.training/blog/2026-01-02-adversarial-token-integration/ Fri, 02 Jan 2026 00:00:00 GMT Most token integrations assume transfer amount fidelity and standard return behavior. Production incidents show the opposite: token behavior is adversarial from the integrator’s perspective. https://crypto.training/blog/2026-01-01-erc4337-auditor-threat-model/ https://crypto.training/blog/2026-01-01-erc4337-auditor-threat-model/ Thu, 01 Jan 2026 00:00:00 GMT Account abstraction audits fail when they treat UserOperation validation as a normal transfer flow. This post maps the real trust boundaries for EntryPoint systems. https://crypto.training/blog/2025-12-31-signing-infrastructure-threat-model/ https://crypto.training/blog/2025-12-31-signing-infrastructure-threat-model/ Wed, 31 Dec 2025 00:00:00 GMT Most major losses now happen above the cryptographic primitive layer. This post maps where signing pipelines fail and how to harden them end-to-end. https://crypto.training/blog/2025-12-30-l2-oracle-failure-modes/ https://crypto.training/blog/2025-12-30-l2-oracle-failure-modes/ Tue, 30 Dec 2025 00:00:00 GMT On L2s, oracle correctness is not only about data source quality. Sequencer liveness, update lag, and fallback design can convert good feeds into bad liquidations. https://crypto.training/blog/2025-12-29-reentrancy-2026-callback-surfaces/ https://crypto.training/blog/2025-12-29-reentrancy-2026-callback-surfaces/ Mon, 29 Dec 2025 00:00:00 GMT Reentrancy is not just withdraw-before-update. Token standard callbacks, cross-function paths, and read-only windows still create severe failures. https://crypto.training/blog/2025-12-28-deployment-layer-security-cpimp/ https://crypto.training/blog/2025-12-28-deployment-layer-security-cpimp/ Sun, 28 Dec 2025 00:00:00 GMT Many teams secure implementation code and still lose funds during deployment. This post maps the proxy deploy-init race, CPIMP persistence techniques, and practical hardening. https://crypto.training/blog/2025-12-27-compositional-rounding-attacks/ https://crypto.training/blog/2025-12-27-compositional-rounding-attacks/ Sat, 27 Dec 2025 00:00:00 GMT Rounding bugs are usually not single-line bugs. They emerge when multiple safe-looking operations compose into a leak loop. This post shows how to model, test, and patch that class of failures. https://crypto.training/blog/2025-12-26-mellow-vs-roles-curator-calldata-checks/ https://crypto.training/blog/2025-12-26-mellow-vs-roles-curator-calldata-checks/ Fri, 26 Dec 2025 00:00:00 GMT A code-level comparison of how Mellow Flexible Vaults and Zodiac Roles Modifier authorize curator transactions: allowlists, calldata constraints, allowances, adapters, and failure semantics. https://crypto.training/blog/2025-12-22-veda-vs-mellow/ https://crypto.training/blog/2025-12-22-veda-vs-mellow/ Mon, 22 Dec 2025 00:00:00 GMT A code-level comparison of Veda BoringVault and Mellow Flexible Vaults: module boundaries, async flows, curator calldata controls, oracle models, and security assumptions. https://crypto.training/blog/2025-12-14-erc-7540/ https://crypto.training/blog/2025-12-14-erc-7540/ Sun, 14 Dec 2025 00:00:00 GMT A code-first analysis of ERC-7540, ERC-7575, and ERC-7887: what the standards require, how production systems actually implement async flows, where designs diverge, recurring vulnerability patterns, and how Mellow compares as an async vault architecture outside strict ERC-7540. https://crypto.training/blog/2025-11-09-fluid/ https://crypto.training/blog/2025-11-09-fluid/ Sun, 09 Nov 2025 00:00:00 GMT A code-informed, security-first walkthrough of Fluid's Liquidity Layer, DEX v2 (D3/D4), and Money Market: architecture, user stories, fund flows, invariants, and vulnerability hunting playbooks. https://crypto.training/blog/2025-10-04-kiln-defi/ https://crypto.training/blog/2025-10-04-kiln-defi/ Sat, 04 Oct 2025 00:00:00 GMT A code-level breakdown of Kiln DeFi vault architecture, connector model, fund flows, user deposit and withdrawal journeys, role/governance controls, and practical risk analysis. https://crypto.training/blog/2025-09-23-safe-evolution/ https://crypto.training/blog/2025-09-23-safe-evolution/ Tue, 23 Sep 2025 00:00:00 GMT A deep, code-first analysis of Safe version evolution from v1.0.0 to v1.5.0: what changed, why it changed, what risks were introduced, and how audits and fixes shifted the security posture over time. https://crypto.training/blog/2025-09-08-ecdsa/ https://crypto.training/blog/2025-09-08-ecdsa/ Mon, 08 Sep 2025 00:00:00 GMT A beginner-friendly walkthrough of ECDSA on secp256k1: private key, public key, signing, verification, and why it is one-way. Includes animated visuals for point multiplication.