⚠️ EDUCATIONAL PROTOTYPE / TEST PROJECTThis is a research prototype designed to test the capabilities of LLMs (Gemini 3.0) in analyzing blockchain market data. It is NOT a production-ready trading bot.
This project is an experimental intelligent agent that autonomously monitors cryptocurrency markets and uses Google Gemini AI to generate insights.
Current Capabilities:
- ✅ Market Data: Fully implemented using CoinGecko API (Global market cap, trends, top tokens).
- ✅ AI Analysis: Fully implemented using Google Gemini 3.0 Flash. It receives market data and generates textual strategies/alerts.
- 🚧 Exchange Integration (CCXT): Partially Implemented.
- The project attempts to fetch public order book data from Binance, Coinbase, and Kraken using
ccxt. - No execution logic: The agent cannot place trades. It only reads prices and calculates theoretical arbitrage spreads.
- This module is currently in a "read-only" prototype state.
- The project attempts to fetch public order book data from Binance, Coinbase, and Kraken using
- Market Analysis: Real-time tracking of Global Market Cap, BTC Dominance using CoinGecko.
- Theoretical Arbitrage Scanner: Compares public order books to find price discrepancies > 1% (Simulation only).
- AI Strategy Generation: Generates "Buy/Hold/Sell" opinions based on aggregated data.
- Console Dashboard: A CLI interface that prints cycle reports.
- Core Logic:
src/core/agent.ts: Main loop.src/mcp/coingecko.ts: Working CoinGecko integration.src/llm/gemini.ts: Working Gemini AI integration.
- Experimental/Prototype:
src/services/exchange-connectors.ts: Wrapper aroundccxt.src/services/arbitrage-finder.ts: Logic to compare prices. Note: Real-world arbitrage requires much lower latency than this Node.js app can provide.
-
Clone the repository
git clone https://github.com/Vision70s/tradeBot.git cd tradeBot -
Install dependencies
npm install
-
Configure Environment
cp .env.example .env
- Required:
GEMINI_API_KEY(from Google AI Studio). - Optional:
COINGECKO_API_KEY.
- Required:
To start the agent's monitoring loop:
npm run devnpm run docker:logs
## 📝 Disclaimer
This software is for educational purposes only. Do not use for real financial trading. The authors are not responsible for any financial losses.