An AI-driven trading framework for OpenClaw that automates crypto and prediction market execution with built-in risk management.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-trading-bot
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 agent-trading-bot 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 Agent Trading Bot?
The Agent Trading Bot is a comprehensive framework designed for the Openclaw Skills ecosystem, enabling developers to build, test, and deploy autonomous trading agents. It bridges the gap between AI reasoning and financial execution by providing secure connections to major crypto exchanges like Binance and Hyperliquid, as well as prediction markets like Polymarket.
This skill prioritizes safety with mandatory paper trading modes and robust risk parameters. By leveraging Openclaw Skills, users can transform high-level trading ideas into executable strategies with granular control over position sizing and stop-loss mechanisms, ensuring that autonomous agents operate within strictly defined financial guardrails.
Agent Trading Bot Use Cases
- Automated dollar-cost averaging (DCA) for long-term crypto accumulation.
- Executing high-frequency grid trading strategies in volatile markets.
- Arbitraging funding rates between perpetual futures platforms.
- Developing autonomous agents that trade based on prediction market sentiment.
- Backtesting custom RSI or momentum-based strategies before live deployment.
How Agent Trading Bot Works
- Connectivity Check: The agent verifies API credentials and exchange status using the status command to ensure a stable link.
- Strategy Definition: Users select a built-in strategy (like DCA or Grid) or define a custom JSON strategy template.
- Backtesting & Simulation: The bot runs the strategy against historical data or in a real-time paper trading environment to validate performance.
- Risk Assessment: The framework checks the portfolio against predefined safety limits such as maximum drawdown and position size.
- Live Execution: Upon explicit user confirmation, the bot executes trades on live exchanges while monitoring for kill-switch triggers.
Agent Trading Bot Setup
Configure your exchange credentials as environment variables to keep them secure within the Openclaw Skills environment:
export BINANCE_API_KEY="your-key"
export BINANCE_API_SECRET="your-secret"
Initialize the trading configuration file and check connectivity:
# Check connectivity
python3 scripts/agent-trading-bot.py status
Define risk limits in ~/.openclaw/trading-config.json to ensure the agent operates within your safety boundaries, including stop-loss percentages and daily trade limits.
Agent Trading Bot Data Schema & Taxonomy
The skill maintains a structured approach to trading data and configuration within the Openclaw Skills architecture:
| Data Type | Location | Description |
|---|---|---|
| Configuration | ~/.openclaw/trading-config.json |
Global risk limits and allowed trading pairs. |
| Strategies | strategies/*.json |
JSON definitions for entry/exit logic and indicators. |
| Logs | logs/trading/ |
Detailed execution logs, including reasoning for every trade. |
| Session State | Internal Memory | Tracks current unrealized P&L and drawdown status. |
Agent Trading Bot Advanced Features
- Emergency Kill Switch: Instantly close all open positions and cancel pending orders across all connected exchanges with a single command.
- Custom Strategy Engine: Define complex entry and exit conditions using a modular JSON schema for advanced technical analysis.
- Multi-Exchange Support: Simultaneous management of positions across Binance, Hyperliquid, Bluefin, and prediction markets.
- Real-time Risk Dashboard: Comprehensive visualization of margin utilization, portfolio exposure, and drawdown metrics.
- Autonomous Reasoning: Every trade is logged with the underlying AI reasoning, providing full auditability of the decision-making process.
SKILL.md
Loading
Related Openclaw Skills
Security Hardener
A comprehensive security auditing and auto-remediation tool designed to harden OpenClaw instances against vulnerabilities and misconfigurations.
Cognitive Loop
A high-autonomy agent architecture that implements a comprehensive Think-Action-Reflect-Test cycle for sophisticated task management.
Self-Healing Agent
A powerful auto-repair and health monitoring system designed to keep Openclaw Skills running smoothly by detecting and fixing failures in real-time.
Car Wash Promo Video
Automates the production of professional marketing videos for car washes and detailing studios to drive customer acquisition and membership sign-ups.
asclaude-compact
A sophisticated session optimization tool that implements layered compression to manage AI context memory without consuming additional tokens.
Vision - Omni Image Recognition
A comprehensive image analysis tool for Openclaw Skills that recognizes text, objects, and scenes with automatic translation capabilities.
Featured*
