ChainScore

A way for businesses and users to gauge the health of web3 wallets.

  • 0 Raised
  • 141 Views
  • 0 Judges

Categories

  • HawkHacks Global Category
  • NEAR Sponsored Prize Category
  • Best Use of PropelAuth

Gallery

Description

Links:

StQiLi/ChainScore (github.com)

Inspiration:

Inspired by the growth and utility of AI in Web3 and blockchain technologies 🌐, we wanted to develop something to tackle the problem of incentivizing more users (even casual individuals or Web2 business entities) to utilize the NEAR protocol. Through our idea generation process, we recognized that there was already an abundant saturation in the current market with NFTs, smart contracts, and cryptocurrencies. Our project ChainScore looks to motivate users (even those without prior knowledge and currently not in the Web3 ecosystem) to have a source of immediate truth 🔍.

What it does:

ChainScore isn’t just a means to get wallet history, but a protocol-wide foundation for NEAR to offer an accessible and centralized manner to aggregate, quantify, and utilize decentralized data. By implementing one of the most vital aspects of the financial sector in a way we haven’t seen others do before, the undercollateralized loan 💸.

ChainScore utilizes the NEAR RPC along with OpenAI’s most recent GPT-4o, combined with heavily engineered prompting, to look up and evaluate the health of online wallets based on our own developed evaluation scheme based on transaction history. By providing the requester or user a detailed analysis of the account 📊,

Challenges:

As a team, none of us had ever developed on Web3 or had any knowledge of blockchain before HawkHacks. The biggest challenges we ran into were mainly due to the developer ecosystem. Looking through the documentation, numerous APIs, and learning blockchain all throughout, we landed on NEAR RPC. Understanding the infrastructure and looking through the open-source projects allowed us to build an indexer and land on a feasible solution for our MVP 🛠️.

Next Steps:

At the moment, our model does not have a concurrency built in and is one of the next steps we’ve looked towards improving on. In other next steps, we wish to make our project compatible for all other Layer 1 infrastructure as well and are looking into creating a platform where more functionalities can be implemented into the chain.⛓️ 


Attachments