This guide shows you how to setup the kelp bot using the sell strategy. We'll configure it to make a market for native tokens we plan to sell in an upcoming hypothetical ICO. This strategy creates sell offers based on a reference price with a pre-specified liquidity depth.
In our example the tokens are priced against XLM (native). XLM is the native currency in the Stellar network and acts as a bridge (i.e. facilitator) in transactions that involve two different assets. Therefore, it has less counterparty risk. Most assets have a liquid market against XLM and by pricing your asset against XLM you are opening up your asset to be traded against any other asset on the network via path payments .
First, go through the Account Setup guide to set up your Stellar accounts and the necessary configuration file, trader.cfg. In the sell strategy the bot is programmed to sell ASSET_CODE_A.
Download the pre-compiled binaries for kelp for your platform from the Github Releases Page. If you have downloaded the correct version for your platform you can run it directly.
Use the sample configuration file for the sell strategy as a template. We will walkthrough the configuration parameters below.
Sell requires two price feeds, A and B, and it computes its final trading price by dividing A by B.
To give COUPON a stable price against USD, we're going to connect feed A to the priceFeed from Kraken, the popular custodial exchange. We set the DATA_TYPE_A to "exchange" and DATA_FEED_A_URL to "kraken/XXLM/ZUSD". This points our bot to Kraken's price for 1 XLM, quoted in USD.
By design, we always want to price our stablecoin at Kraken's XLM/USD price. Since that's already what we're getting in feed A, and final price is A/B, we just set price feed B to 1.0. We do this by setting DATA_TYPE_B to "fixed" and DATA_FEED_B_URL to "1.0".
For the purposes of this walkthrough, we set the PRICE_TOLERANCE and AMOUNT_TOLERANCE thresholds for our bot to be as small as possible, because we want it to make a lot of trades. We set the value for both fields to 0.001 which means that a 0.1% change in price or amount will trigger the bot to refresh its orders. In practice, you should set any value you're comfortable with.
AMOUNT_OF_A_BASE allows you to scale the order sizes set in the next section of the configuration. Trade amounts are specified in units of the base asset.
A level defines a layer that sits in the orderbook. Each level has an AMOUNT and a SPREAD as part of its configuration. The bot creates mirrored orders on both the buy side and the sell side for each level configured.
AMOUNT_OF_A_BASE allows you to scale the order size levels explained below. Trade amounts are specified in units of the base asset (i.e. ASSET_CODE_A).
- AMOUNT: specifies the order size in multiples of the base unit described above. This
AMOUNTis multiplied by theAMOUNT_OF_A_BASEfield to give the final amount. The amount for the quote asset is derived using this value and the computed price at the indicatedSPREADlevel. - SPREAD: represents the distance from the mid price as a percentage specified as a decimal number (0 < spread < 1.00). The bid/ask spread will be 2x what is specified at each level in the config.
Assuming your botConfig is called trader.cfg and your strategy config is called sell.cfg, you can run kelp with the following command:
kelp trade --botConf ./path/trader.cfg --strategy sell --stratConf ./path/sell.cfg
You can also play around with the configuration parameters of the sample configuration file for the sell strategy, look at some of the other strategies that are available out-of-the-box or dig into the code and create your own strategy.