Requirements
- Target platform
- OpenClaw
- Install method
- Manual import
- Extraction
- Extract archive
- Prerequisites
- OpenClaw
- Primary doc
- SKILL.md
Run autonomous multi-agent trading competitions on OKX. 5 AI agents compete with real-time market data, evolutionary selection replaces losers daily, exchang...
Run autonomous multi-agent trading competitions on OKX. 5 AI agents compete with real-time market data, evolutionary selection replaces losers daily, exchang...
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run.
Run autonomous multi-agent trading competitions where AI strategies compete against each other with real market data.
5 agents compete simultaneously on OKX Demo (or Live) Every 5 minutes: each agent analyzes candles, indicators, and decides to enter/exit Daily evolution at 07:00: top 2 survive, bottom 2 get replaced by better strategies Exchange-level stop-losses: positions are protected even if the bot crashes 3-layer position sizing protection: unknown assets skipped, 50% equity cap, $500 hard max
βββββββββββββββ β Competition β β runs every 5 minutes β Manager β ββββββββ¬βββββββ β ββββββΌβββββ¬βββββ¬βββββ β β β β β v v v v v Agent1 Agent2 Agent3 Agent4 Agent5 (S&D) (RSI) (S&D) (RSI) (RSI) β v βββββββββββ β OKX API β β real-time prices, order execution βββββββββββ
Supply & Demand β Identifies accumulation/distribution zones, enters on retests RSI Mean Reversion β Overbought/oversold with EMA trend filter RSI Fast Confirm β 1-bar bounce confirmation with tight EMA alignment RSI Trend Filter β Long-period EMA filter with standard RSI levels
LayerProtectionWhat Happens1CT_VALS lookupUnknown asset β SKIP (never enter blind)250% equity capMax position = half of agent's capital3$500 hard capAbsolute maximum notional per trade4Exchange SLStop-loss on OKX itself (survives bot crash)5GuardianWindows Task checks every minute, restarts if dead
OKX account (Demo or Live) API Key + Secret + Passphrase Node.js 18+
Copy bybit-trading/ folder to your workspace Create .secrets/okx.env: OKX_API_KEY=your_key OKX_API_SECRET=your_secret OKX_PASSPHRASE=your_passphrase Edit agents_config.json with your preferred strategies and assets Run: node competition_manager_okx.js
Add a Guardian cron or Windows Task that checks competition_log_okx.txt freshness every minute. If log hasn't updated in 6 minutes β kill and restart.
Daily at 07:00: π₯ Rank 1 β Survives (strategy + assets) π₯ Rank 2 β Survives (strategy + assets) π₯ Rank 3 β Strategy stays, assets rotate to better performers 4οΈβ£ Rank 4 β Fully replaced from best_strategies pool 5οΈβ£ Rank 5 β Fully replaced from best_strategies pool Criteria for new agents: PF > 1.5, DD < 25%, min 5 backtested trades.
Before going live with real money: 7 days continuous run without crashes P&L > 15% on demo 30+ trades for top 3 agents Max drawdown < 10% 0 sizing bugs Exchange-level SL verified on all new positions
Node.js + OKX REST API Technical indicators (EMA, RSI, ATR, Supply/Demand zones) Evolutionary selection algorithm Real-time Telegram alerts
Agent frameworks, memory systems, reasoning layers, and model-native orchestration.
Largest current source with strong distribution and engagement signals.