How does the system handle trading risks?

It includes a mandatory Risk Agent that runs before trades, plus configurable circuit breakers such as MAX_LOSS_USD, MAX_GAIN_USD, and minimum balance requirements.

Can I switch between different AI models?

Yes, the system uses a ModelFactory interface that supports various providers including Anthropic (Claude), OpenAI, DeepSeek, Groq, Gemini, and local models via Ollama.

Which exchanges are compatible with this system?

The system natively supports Hyperliquid (perpetual DEX), Solana via BirdEye for spot tokens, and Extended Exchange (X10) for StarkNet-based perpetuals.

What is the recommended development environment?

The repository is built for Python 3.10.9. It follows a modular architecture where each agent is a standalone script under 800 lines of code for easier debugging.

What is the RBI agent?

The Research-Based Inference (RBI) agent is a specialized tool that takes a strategy idea from a source like a YouTube URL or PDF and uses AI to generate and run a backtest using the backtesting.py library.

Moon Dev AI Trading Agents

Name: Moon Dev AI Trading Agents
Author: Activer007

byActiver007

0•

Ciencia de Datos y ML

Orchestrates 48+ specialized AI agents for autonomous cryptocurrency trading and market analysis across multiple exchanges.

This skill provides expert knowledge for interacting with Moon Dev's experimental AI trading system, a modular framework designed to manage autonomous trading workflows. It enables developers to deploy and customize over 48 specialized agents that handle tasks ranging from market sentiment analysis to high-frequency execution on platforms like Hyperliquid and Solana. By leveraging a unified ModelFactory, the skill allows for seamless integration with leading LLMs while maintaining strict risk management protocols and backtesting capabilities through the unique Research-Based Inference (RBI) agent.

Características Principales

01Orchestration of 48+ specialized AI agents for trading, research, and analysis

02Multi-exchange support for Hyperliquid, Solana (BirdEye), and Extended Exchange (X10)

03ModelFactory abstraction to switch between Claude, GPT-4, DeepSeek, Groq, and Gemini

04Integrated risk management system with circuit breakers and AI-driven position confirmation

05Automated backtesting via the RBI agent using strategy logic extracted from videos or PDFs

060 GitHub stars

Casos de Uso

01Deploying autonomous AI-driven trading strategies across perpetual and spot markets

02Backtesting complex trading ideas using AI to generate executable Python code from research materials

03Automating real-time market sentiment and whale activity monitoring for crypto assets

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add activer007/ordinary-claude-skills moon-dev-trading-agents

For use in Claude.ai and ChatGPT

Download Skill