case study · automated trading

SOL Trading Bot

Rules-based, on Solana — market patterns call the shots, not emotion. Copies known wallets' trades, prices every entry and exit through Jupiter, and logs the result.

Renaissance-style illustration of two figures studying a glowing web3 trading dashboard, in Prometrifi's dithered Digital Renaissance style.

the gap

Copy-trading by hand means staring at wallets, not living your life.

Following a handful of wallets means watching for their entries and exits in real time, then moving fast enough to still get a fair price. Miss the window by a few minutes and the trade you're copying isn't the trade you actually get. That's not a discipline problem — it's a job for something that doesn't sleep.

how it runs

A 15-minute cycle, the same logic every time.

Track

Every 15 minutes, active wallets and their open trades are pulled fresh, and new transactions are attached to the wallet that made them.

Classify

New transactions are merged against open positions and sorted into buys and sells — the same rule, applied every cycle.

Price & route

Live pricing is pulled and each action is routed to open a new copy trade or close and exit an existing one.

Log

The result is written to the trade log and each wallet's cursor is updated, so the next cycle picks up exactly where this one left off.

built with

The actual stack.

n8n workflow automation Nansen wallet data Jupiter aggregator pricing Postgres trade log

where it stands

Polling every 15 minutes today — live, with no open positions yet.

the architecture

One pipeline, 4 stages.

01 Track Active wallets and open trades, pulled fresh.

→

02 Classify New transactions sorted into buys and sells.

→

03 Price & route Live pricing, routed to open or close a position.

→

04 Log Result recorded, cursor updated for next cycle.

proof of work — the pipeline

The n8n canvas, in full.

n8n workflow canvas: a 15-minute schedule trigger fetching active wallets and open copy trades, merging and classifying buys and sells, routing by action through Jupiter buy and sell pricing, and updating open or closed copy trades and wallet cursors. — Every 15 minutes: fetch active wallets and open trades, classify each transaction, price it through Jupiter, and route it to open or close a copy trade.

proof of work — sample output

The daily digest.

Slack message from Web3 Dashboard Bot: Nansen SOL Copy-Trade digest showing open positions, unrealized profit and loss, closed trades, win rate, average return, and realized profit and loss. — A daily digest posted straight to Slack — open positions, win rate, and realized P/L at a glance. Currently reading zero across the board, which is exactly what an honest early-stage log should show.

let's work together

Have a project in mind?

hello@prometrifi.co
← back to all projects