Build an XRPL Token Screener in About 50 Lines
There are 20,268 tokens issued on the XRP Ledger and only 1,543 of them traded today. A screener is how you tell those two groups apart. Here is a complete one. No dependencies, runs on any Node 18+. The whole thing const API = ' https://api.xrpl.to/v1 ' ; async function get ( path , attempt = 1 ) { const r = await fetch ( API + path ); if ( r . ok ) return r . json (); if (( r . status === 429…
The article provides a guide on how to create an XRPL token screener with only 50 lines of code. The XRPL Ledger currently hosts 20,268 tokens, but only 1,543 of them traded today. A screener helps distinguish between these two groups. The article presents a complete code example that runs on any Node 18+ and does not require any dependencies. The code is divided into three main functions: get, score, and screen.
The get function fetches data from the specified API endpoint and handles potential errors, such as rate limits or server issues. It returns the JSON data received from the API. The score function calculates two metrics for each token: perTrader and stickiness. perTrader measures the volume per unique trader, indicating low values when few accounts are moving large amounts, which could signal wash trading.
Stickiness measures the ratio of current holders to total trustlines ever opened, indicating low values when people open a line, trade, and then leave. The screen function fetches the top 100 tokens sorted by volume, filters out tokens with less than 1000 holders or fewer than 20 unique traders in the last 24 hours, and then maps the filtered tokens to an object containing the name, issuer, score, holders, and volume.
Finally, it filters out tokens with stickiness below 0.15, sorts the remaining tokens by perTrader in ascending order (meaning low volume per trader, indicating many accounts each trading modestly), and logs the top 15 tokens along with their respective metrics.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.