Run an automated perpetual futures bot on Hyperliquid in paper mode
Backtest the 4-hour trend and 5-minute scalp signals against historical data
Plug in Telegram alerts for entries, exits, and drawdown stops
Tweak the amplifier and veto logic to fit a different risk profile
Needs Python 3.10+, a Hyperliquid API wallet, and you should start in PAPER mode before going LIVE because real money is at risk.
WhuffoBot is an automated trading bot for the Hyperliquid decentralized exchange, written in Python. It opens perpetual futures positions on its own. Every 15 minutes it scans all available pairs on Hyperliquid, scores them, and acts on whichever one produces the strongest signal at the moment. The README also lists two backtests: a 2024 bull-run window that returned 84.6%, and a 90-day sideways or bear period that returned 13.6%. The bot runs two strategies at the same time. The primary one is a 4-hour trend strategy. It opens a long when a fast EMA crosses above a slow EMA, RSI is at least 50, and ADX is at least 25, and the mirror conditions for a short. The secondary one is a 5-minute scalp strategy that fires only on a tight combination of a Bollinger Band squeeze breakout, a volume spike, an extreme RSI reading, and price deviation from VWAP. On top of these, amplifiers (rising open interest, Bitcoin moving the same way, favorable funding, a nearby liquidation cascade, or the top five Hyperliquid whales positioned the same way) each bump the position multiplier by one step, with the multiplier ranging from 2x to 7x for trend trades and 5x to 10x for scalps. Veto conditions, such as adverse funding or a sharp Bitcoin move against the trade, cancel the signal outright. Risk management is fixed. Stop loss sits at 2%, take profit at 4%. When the first target hits, half the position closes, the stop moves to breakeven, and a trailing stop carries the rest. Scalps are force-closed after 60 minutes. A reverse EMA cross closes the rest of the trade. If account drawdown reaches 25%, the bot stops opening new positions. A regime detector watches Bitcoin's EMA200 and ADX to adjust position sizing and block shorts in bull runs, and a learner system tracks win rate per signal type and drops signals that have done poorly. Setup needs Python 3.10 or newer. You copy config.example.py to config.py, fill in a Hyperliquid API wallet key and your main address, and run python main.py. The README strongly suggests starting in PAPER mode before going LIVE. Public Binance endpoints supply OHLCV, open interest, and funding data, with no Binance account needed. Telegram alerts are optional, and launchd or systemd snippets cover auto-start. License is MIT, with a disclaimer that this is not financial advice and trading carries real risk.
Generated 2026-05-22 · Model: sonnet-4-6 · Verify against the repo before relying on details.