Lending Protocol Subgraph Standard

This repository contains a standard schema for lending protocols.

Types of Lending Protocols

The existing decentralized lending protocol all have slightly different implementations of lending protocols. A non-exhaustive list is provided below.

• Aave & Comp - Pooled. Global risk. Multi-asset. Borrow assets that are deposited.
• Maker & Abracadabra - Collateralized Debt Positions (CDPs). Isolated. Borrow a single synthetic asset minted, such as DAI or MIM.
• Synthetix - Pooled. Global risk. Single asset as collateral (SNX). sUSD and many other synthetic assets minted.
• RAI, Liquity - CDP, single asset as collateral, ETH. Borrow a single synthetic asset, RAI or LUSD.

Generalizing a lending protocol

Each protocol implements their own functions, events, token minting and token transfers. This means we have to make a very general schema to support all of these possibilities. Below is a general framework:

• Deposit - Deposit collateral into a protocol.
• Withdraw - Withdraw collateral from a protocol.
• Borrow - Borrow an asset from a protocol.
• Repay - Repay a borrowed asset from a protocol.
• Liquidate - Liquidate a user for profit.

Config and Naming of Subgraphs

• Each Subgraph is located in it's own folder within /subgraphs.
• Each Subgraph must adopt the schema.graphql file as the schema. This file is essentially the standard, and thus it should be the only one used in this repository. However, forks of this repo can create new schemas in each Subgraph folder.
• Each Subgraph is named by it's best known protocol and version.
  • For a clear example, let's consider Uniswap. We would name the folder uniswap-v2 and inside of it would be a config file for sushiswap, because sushiswap is a fork of Uniswap V2.
• In the config folder of each Subgraph, it will contain all forks and networks that exist for a base set of Smart Contracts.
• Each config folder will contain json files with the config. Then mustache can be used in combination with node scripts to deploy all networks and forks of a single Subgraph.

Priority of implementing the Subgraph

• Events and Counts first - as they are straight forward. Test the indexing speed here (and at the next steps).
• Then do all the required stuff - real asset amounts held by accounts and protocols.
• Then USD because it will be complicated and it will be nice to have before the Lifetime aggregation.
• Then do Lifetime aggregation amounts.