Cross Credit

Demo

About

The problem it solves Problem it solving : 1. **Cross-Chain Lending Without Complexity** Traditionally, borrowing against your crypto on another chain requires: * Using bridges (which are slow and risky) * Manually moving assets between chains * Repeating KYC or approvals across dApps **Cross Credit simplifies this** by letting users deposit collateral on one chain and borrow from another — instantly, securely, and trustlessly — using **Chainlink CCIP**. --- 2. **No On-Chain Reputation System** DeFi lacks a standardized way to measure a user’s trust or behavior. Every loan is treated as high risk, leading to: * Overcollateralization (150%+) * High interest rates * Limited access for new users **Cross Credit introduces the DeFi Passport**, a soulbound NFT that reflects real user activity: * Borrowing history * Liquidation risk * Wallet activity * DAO involvement This enables **dynamic interest rates, trust scores, and personalized loan terms** — similar to how credit scores work in TradFi. --- 3. **Lack of Transparency in DeFi Lending** Current platforms calculate risk in a black box — users don’t know why they get poor terms. With Cross Credit: * You **see your credit score** and how it was calculated. * Your **DeFi Passport is reusable** across apps. * You get **badges** for good behavior (e.g., "Never Liquidated", "DAO Voter"). This adds **clarity and gamification** to the lending process. --- 4. **Security and Trust Concerns in Cross-Chain Systems** Bridges and multi-chain protocols are common targets of hacks and rug pulls. **Cross Credit uses Chainlink’s secure infrastructure** (Functions + CCIP), ensuring: * Safe API calls * Tamper-proof credit scoring * Verified cross-chain execution This brings **bank-grade security to DeFi loans**. --- 5. **No Incentives for Responsible Users** In most lending protocols: * Everyone pays the same rate * Good behavior isn't rewarded **Cross Credit changes this**: * **Lower interest** for high-score users * **Higher limits** and faster approvals * **Token/NFT rewards** for consistent performance It turns borrowing into a **long-term, reputation-based experience**, not just a one-time transaction. --- 6. **DAO and Ecosystem Integration** DAOs currently have no easy way to verify user credibility when: * Granting funds * Allocating votes * Selecting contributors **Cross Credit enables DAOs** to: * Filter users by score * Offer credit-based access control * Reward verified contributors

How it's made

Challenges we ran into Challenges I Ran Into 1. **Transaction Reverts on Deposit** One of the earliest bugs occurred during the deposit flow on Sepolia testnet. Transactions would revert without a clear error message, even though the smart contract logic was correct. **What caused it:** * The user wallet didn’t have WETH approval. * Contract expected WETH but received native ETH. **How I solved it:** * Added a utility to auto-wrap ETH to WETH if needed. * Implemented proper ERC-20 allowance checks before `transferFrom`. * Improved error handling in the frontend to show user-friendly messages. --- 2. **Adding WETH as a Supported Token** When running the script to register WETH as a supported token, it failed silently at first — even though logs said it was successful. **Root cause:** * The contract didn't store the token due to a missing storage mapping update in the contract. **Fix:** * Added detailed `console.log()` outputs in Hardhat scripts. * Refactored the `addSupportedToken()` method to emit events and verify on-chain updates post-transaction. --- 3. **Chainlink CCIP Setup** Setting up secure cross-chain messaging between the Depositor and Minter contracts was tricky. **Challenges:** * Each chain has its own CCIP router address. * Misconfigured gas limits and fee tokens broke the message relay. **Solution:** * Created a centralized config file for Chainlink router and DON IDs per chain. * Added retry logic and error events to debug relay issues. * Consulted Chainlink docs and community for proper testnet fee token usage. --- 4. **Smart Credit Score Calculation** Creating a fair and accurate credit scoring algorithm from wallet activity was difficult, especially due to: * Inconsistent on-chain history formats * Data sparsity for new users * Score inflation from wash trading **How I overcame it:** * Fetched wallet data via Covalent and Alchemy APIs. * Applied ML scoring models trained on real user behaviors. * Added penalty weights for liquidation and suspicious activity. * Used Chainlink Functions to fetch off-chain scores on demand. ---