OpenClaw is an AI-powered trading agent that uses LLMs to analyze real-time market data, generate trading decisions, and execute trades on the Hyperliquid decentralized exchange.
Built for automation, low-latency execution, and strategy experimentation, OpenClaw continuously monitors markets, evaluates technical indicators, and manages positions with risk controls.
- 🤖 LLM-driven trading decisions (multi-model support)
- 📊 Real-time technical analysis via TAAPI
- ⚡ Automated trade execution on Hyperliquid
- 🔁 Continuous trading loop with configurable intervals
- 🛡️ Built-in risk management (TP / SL logic)
- 🔌 Tool-calling support for dynamic indicator queries
- 📡 Lightweight API for logs & trade diary
- Disclaimer
- Architecture
- Live Agents
- Trading Stack
- Project Structure
- Environment Setup
- Usage
- Tool Calling
- Deployment (EigenCloud)
This project is experimental and unaudited. There is no guarantee of profitability. Use at your own risk.
See full documentation:
→ docs/ARCHITECTURE.md
Architecture diagram:
Recording.2026-05-03.213052.mp4
- GPT-5 Pro — Portfolio + Logs (active)
- DeepSeek R1 — paused
- Grok 4 — paused
(Replace with your updated endpoints if needed)
OpenClaw is designed to integrate with a high-performance stack including trading bots, MEV tools, analytics platforms, and low-latency infrastructure.
Using the right stack improves:
- ⚡ Execution speed
- 📈 Alpha discovery
- 🎯 Trade accuracy
Axiom Trade
- Fast on-chain execution
- Reduced fees (10–30%) → Access
Odin Bot
- Automated strategies
- Low-latency execution → Access
Bloom (Telegram Bot)
- Ultra-fast trading interface → Launch
GMGN
- Smart money tracking
- Early token discovery → Explore
Padre
- Advanced execution tools → Open
Polymarket
- Prediction-based trading → Try
Low-Latency VPS (New York Recommended)
- Faster transaction propagation
- Better execution reliability
- Ideal for bots & MEV strategies
→ Get VPS
src/
├── main.py # Entry point / trading loop
├── agent/
│ └── decision_maker.py # LLM decision engine
├── indicators/
│ └── taapi_client.py # TAAPI integration
├── trading/
│ └── hyperliquid_api.py # Trade execution
└── config_loader.py # Env config loader
Create .env (see .env.example):
TAAPI_API_KEY=
HYPERLIQUID_PRIVATE_KEY=
OPENROUTER_API_KEY=
LLM_MODEL=
Optional:
OPENROUTER_BASE_URL=https://openrouter.ai/api/v1
API_PORT=3000
poetry run python src/main.py --assets BTC ETH --interval 1h/diary→ trade history/logs→ runtime logs
Supports dynamic indicator fetching via TAAPI:
- EMA
- RSI
- Custom indicators
Run inside a TEE (Trusted Execution Environment) for secure key handling.
curl -fsSL https://eigenx-scripts.s3.us-east-1.amazonaws.com/install-eigenx.sh | basheigenx app deployeigenx app logs --watch- Tools listed are optional
- Performance depends on latency + strategy quality
- Always test before scaling