Best Practices for Detecting and Preventing Risks in Web3 Development
summary
Hypernative claims continuous real-time monitoring that flags preparatory behavior so attacks are caught before the first exploit transaction — the company reports detecting 98% of historical exploits early, with the Jimbos hack flagged 7 minutes beforehand and Raft Finance ~15 minutes prior. The system uses a watch list, contract creation ML scoring, on‑chain simulation/fuzzing and off‑chain signals, surfaces risk insights (vulnerabilities, governance changes, large transfers) and can trigger automated responses to pause or limit protocol activity.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 1:07Web3 security and monitoring
- 3:19Limitations of audit-only security
- 5:07Real-time security solutions
- 5:51Hyper Native’s attack detection findings
- 7:36Setting up a watch list
- 9:25Risk insights and detection methods
- 10:30Covering all attack vectors
- 13:45Case study: Jimbos protocol hack
- 16:20Identifying attacker preparation
- 19:50Selecting a monitoring solution
- 20:55Defining policies and responses
- 24:13Custom agents for security
- 26:23Security for normal users
- 28:57Hyper Native’s mission
- 31:48Layered security strategy
- 32:31Closing remarks
related talks
12mReal-Time Security in Solana Ecosystem
43mCybersecurity and Global Collaboration for Securing the Web 3.0 Ecosystem
01h 33mWhite Hat Panel: DeFi Exploits
06mRuntime Threat Detection and Automated Response
20mA cat-and-mouse game: how to frontrun a transaction in the future
22mA Multi-Layer Dynamic Security Framework for DeFi Smart Contracts
31mCybersecurity in Web3
17mUsing AI to anticipate risk