An autonomous framework for running multi-agent AI trading competitions with real-time market data and evolutionary strategy selection.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install trading-tournament
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install trading-tournament using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Trading Tournament AI Agent?
The Trading Tournament skill is a robust, battle-tested framework designed for algorithmic trading research and execution. It enables users of Openclaw Skills to deploy five independent AI agents that compete simultaneously using real-time market data from the OKX exchange. By leveraging diverse strategies like Supply and Demand zones and RSI variants, the system creates a competitive environment where only the most profitable strategies survive.
The system is built with a heavy emphasis on capital preservation, featuring multiple layers of protection including exchange-level stop-losses and strict position sizing. This skill transforms a standard trading setup into a self-evolving ecosystem that identifies and promotes high-performance trading logic through daily evolutionary cycles.
Trading Tournament AI Agent Use Cases
- Developing and stress-testing new algorithmic trading strategies in a live or demo environment.
- Automating portfolio management through a survival of the fittest agent selection process.
- Reducing emotional bias in trading by delegating execution to a multi-agent AI framework.
- Researching the effectiveness of evolutionary algorithms in financial market prediction.
How Trading Tournament AI Agent Works
- The Competition Manager initializes five unique AI agents, each assigned a specific strategy and asset pool.
- Every five minutes, the system pulls real-time candle data and technical indicators via the OKX API.
- Each agent analyzes the data independently to decide on entry, exit, or hold positions based on their specific logic.
- Safety layers validate every trade request, checking against equity caps and hard dollar limits before execution.
- At 07:00 daily, the evolution module ranks agents by performance; top agents survive while underperformers are replaced by new strategies from a backtested pool.
Trading Tournament AI Agent Setup
First, ensure you have Node.js 18+ and an OKX account (Demo or Live).
- Copy the trading folder to your project workspace.
- Configure your credentials in
.secrets/okx.env:
OKX_API_KEY=your_key
OKX_API_SECRET=your_secret
OKX_PASSPHRASE=your_passphrase
- Define your strategies and assets in
agents_config.json. - Launch the tournament:
node competition_manager_okx.js
- Optional: Set up a Guardian cron job or Windows Task to monitor
competition_log_okx.txtand restart the process if it stalls for more than 6 minutes.
Trading Tournament AI Agent Data Schema & Taxonomy
The skill maintains state and logs to ensure transparency and recoverability across competition cycles.
| File | Purpose |
|---|---|
agents_config.json |
Defines agent strategies, assets, and risk parameters. |
okx.env |
Secure storage for exchange API credentials. |
competition_log_okx.txt |
Real-time heartbeat and trade execution logs for the Openclaw Skills framework. |
evolution_history.json |
Records daily performance rankings and strategy rotations. |
Trading Tournament AI Agent Advanced Features
- Evolutionary Logic: Automatically rotates assets for mid-tier performers and fully replaces bottom-tier agents with high-performing backtested strategies.
- Exchange-Level Safety: Implements stop-losses directly on the OKX exchange to protect capital during bot downtime or network issues.
- Multi-Layer Guardian: Uses a watchdog process to ensure 24/7 uptime by restarting the competition manager if logs stop updating.
- Position Sizing Protection: Includes a 3-layer check system featuring unknown asset skipping, a 50% equity cap per agent, and a $500 hard maximum per trade.
SKILL.md
Loading
Related Openclaw Skills
OKX AI Trading Competition Manager
A 24/7 autonomous AI trading system where five agents compete in a daily tournament on OKX to refine strategies.
Hype Scanner (Ari)
An autonomous 24/7 scanner that detects real crypto and stock hype using local AI validation to eliminate market noise.
Auto-Watchdog
A specialized health-monitoring and self-healing skill that ensures continuous uptime for your Openclaw agents by monitoring logs, crons, and system resources.
Anti-Amnesia: Persistent Memory System
A robust state management and persistent memory framework designed to prevent context loss in AI agents across session restarts.
OpenClaw IM Channel Integration
A unified configuration suite to connect OpenClaw AI agents with popular messaging platforms like Telegram, Slack, and Discord.
Tavily Search
An AI-optimized web search skill that delivers clean, relevant web data and deep research capabilities for Openclaw Skills agents.
Featured*
