defiphoenix / erc20-trading-bot-ts

Trading bot for erc20 tokens

Trading Bot

Crypto trading bot which trades erc20 token pairs.

⚠️ Warning:

This project is only a pure hobby project, which came up through my interest in trading strategies and the used tech stack. The bot is not used by me and the reliability and chances of success of the strategies has not been tested in any way.

Prerequisites

• Node and npm/yarn installed
• A websocket (wss) url to a provider is needed. Either register online for one or host a local node.
• A polygon wallet is needed with matic for the fees and the configured stable token (usdc as default).

Getting started

• Open the secrets.json in the src/config folder and fill in the provider url and your private key.
• Run yarn start / npm run start to run the bot.

Options

Configurations can be made in the file src/config/options.json.

Following options are available:

Name	Function	Default
strategy	The strategy which the bot should use	Grid Trading
simulationMode	If true no swapps will be actually made	true
slippage	Change from price in % to buy token	0.1%
maxGwei	The max gwei used for a transaction	80 gwei
gasLimit	Gas limit	250000
coinName	Name of the network coin	Matic
coinDigits	Digits of the network coin	18
stableToken	Address of the stable token	USD Coin address
tradeToken	Address of the traded token	Wrapped Matic address
pairAddress	Address of the stable/traded token pair	USDC/WMATIC pair address
routerAddress	Address of the router/DEX	Quickswap address
refreshTime	Time in milliseconds the price is refreshed	1000
logLevel	Can be either 'info' or 'debug'	debug
serverPort	The port were the server runs on	3000
strategies	Options for the different strategies	-

Strategies

Grid Trading

Options

Name	Function	Default
range	min and max price in which the bot trades	2.05 - 2.65
totalBuyPower	Total amount to invest	1000$
gridMargin	Size of grids in percent	0.5%
rebalance	Currently not implemented	true

How it works

• The bot calculates number of grids in the range.
• The bot calculates the initial grid and buys initial tokens.
• Whenever the price falls one grid the bot buys and when the price goes up one grid the bot sells.
• Note: The total buy power is divided by the number of grids.
• So if the gridMargin is to small the trade amount will be very small.
• If the grid margin is in contrast to big, the bot can't gain profit from small fluctuations.

Scalping

Options

Name	Function	Default
maxTrades	maximum number of trades the bot executes	50 trades
buyPower	Buy power in $ per trade	10$
dropBeforeBuy	Price drop required before buy	0.1%

How it works

• The bot observes the quickswap swaps of the usdc/wmatic pair.
• If wmatic price goes down, the bot buys wmatic for a defined buying power.
• The bot caluculates a wmatic price which will lead to a (minimal) gain.
• If the calculated sell price is reached, the bot swaps back the wmatic for usdc.
• Then the steps are repeated.

Implement your own strategy

Crteating a strategy is very simple. Just follow these steps:

• Create a class for your strategy
• Add options for your strategy if needed
• Add your strategy to the bot so it can be used

Create a strategy class

• First go in the src/strategies folder an create a class for your strategy.
• Implement the Strategy interface. It will provide 3 functions.
• The init(conversion: number) function is called once at startup.
• The priceUpdate(conversion: number, priceChange: number) function is called whenever the price of your trade pair changes.
• The orderLiquidated(order: Order) function is called when one order is executed.

Make a trade

If you want to make a trade in your strategy you first need to create a Order. A order can be created like this:

let order = new Order(OrderType.BUY, buyPower, limit);

After that you can either execute the order directly with

orderBook.executeOrder(order);

or add it to the order book in case you've added an limit to your order.

orderBook.addOrder(buyOrder);

In case the order is added to the order book, after the order is executed the orderLiquidated(order) function in your strategy is called.

In both cases the order will be edited after execution and the amountOut and transactionHash fields will be filled.

A complete example:

let sellOrder = new Order(OrderType.SELL, tradeTokenBalance);
orderBook.executeOrder(sellOrder).then((order: Order) => {
    console.log(`Order executed, got ${order.amountOut} out`);
});

Add the class to the bot

All you need to do is to extend the getStrategy() function and add your strategy.

function getStrategy(): Strategy {
  switch (options.strategy) {
    case 'gridTrading':
      return new GridTrading(options.strategies.gridTrading);
  }
}

The final step is to add strategy options to the options.json file and change the stategy option to your strategies name.