Tech Stack
A breakdown of every technology used in GuardAI and why it was chosen.
Chainlink (Core)
Chainlink CRE
Autonomous workflow engine — runs every 60s, computes risk, writes on-chain
Chainlink Data Feeds
ETH/USD price feed (0x5f4eC3...) — verifiable, manipulation-resistant
CRE is the backbone of GuardAI's autonomy. Without it, the risk monitoring would require a centralized server. With CRE, the entire pipeline is trustless and automated.
Smart Contract Layer
Solidity
^0.8.24
VaultSentinel contract language
Hardhat
Latest
Compile, test, deploy
OpenZeppelin
v5
Ownable, IERC20 — battle-tested primitives
TypeChain
Auto-generated
Type-safe contract bindings for TypeScript
Frontend
Next.js
14 (App Router)
React framework, SSR, routing
wagmi
v2
React hooks for wallet + contract interaction
viem
v2
Low-level Ethereum client (used by wagmi)
Tailwind CSS
v3
Utility-first styling
Framer Motion
Latest
Animations — tab transitions, modals, gauge
GSAP
Latest
Hero section animations
Infrastructure
Tenderly Virtual Testnet
Mainnet fork (Chain ID 9991) — realistic simulation without real funds
DeFiLlama API
TVL data for Lido, Aave, MakerDAO — free, no API key required
Developer Tooling
TypeScript
End-to-end type safety (contracts, scripts, frontend)
ethers.js
Used in Hardhat scripts
dotenv
Environment variable management
ESLint
Frontend linting
Why Chainlink CRE?
Traditional DeFi protection tools rely on:
❌ Centralized bots (single point of failure)
❌ Manual monitoring (human reaction time)
❌ Off-chain oracles with trust assumptions
Chainlink CRE provides:
✅ Decentralized compute — no central server
✅ Native Data Feed integration — verifiable price data
✅ On-chain execution —
eth-transactionstep writes directly to the contract✅ Scheduled automation — cron trigger every 60 seconds
✅ Composable steps — HTTP + JavaScript + on-chain in one workflow
Last updated