AI Agentsai-agentalgorithmic-tradingartificial-intelligencebinancebitcoin
Installation
# Add to your Claude Code skills
git clone https://github.com/asonglin/crypto-liquidity-ai-trading-bot
README.md
Crypto Liquidity AI Trading Bot 🚀
AI trading bot for liquidity detection and algorithmic trading in crypto markets. Detect order book gaps, hidden walls, and liquidity sweeps across exchanges—then act on signals manually or via your own execution layer.
Most crypto trading bots rely only on . Professional traders, however, monitor , because price often moves toward zones where liquidity is concentrated—and away when that liquidity is swept.
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
This project focuses on liquidity-aware trading signals instead of lagging indicators: it detects gaps, walls, and sweeps so you can act on structure, not just price.
Who this is for
This project may be useful for:
Crypto trading firms building in-house liquidity and execution tools
Quant researchers studying order book and market microstructure
Exchanges building surveillance or liquidity analytics
Developers building AI trading agents or signal systems
Commercial use
If you are interested in custom crypto trading bots (liquidity, arbitrage, execution), AI trading signal systems, or liquidity detection algorithms and exchange API integrations:
Win rate > 50% suggests the liquidity-based signals add information over a random baseline.
Profit factor > 1.2 indicates positive expectancy in the simulated period.
Sharpe > 1.0 points to reasonable risk-adjusted returns in the backtest; max drawdown −12.4% is a measure of tail risk in the tested period.
Limitations
Actual results can differ from backtests due to fees, slippage, execution delay, and changing liquidity. Run your own tests and risk checks before any live use.
Large orders and stop-loss clusters sit in the order book before price reaches them. When price sweeps those levels, liquidity is consumed and moves tend to accelerate. This bot identifies those levels and signals sweep events so you can trade with the flow instead of chasing price.
Architecture
Market data (REST/WS)
↓
Order book analyzer
↓
Liquidity detector (gaps, walls, sweeps)
↓
Signal engine
↓
Alerts / optional execution layer
The codebase separates data (modules/, exchange APIs), analysis (trade/ — orderbook, liquidity), and signals/alerts so you can plug in your own execution or research tools.
Quick start
Quick start (Node)
Run the main app with Node:
git clone https://github.com/asonglin/crypto-liquidity-ai-trading-bot.git && cd crypto-liquidity-ai-trading-bot
npm install
Edit config.default.jsonc, then:
node app.js
Advanced research (Python optional)
For research, backtests, or a custom Python wrapper, use a venv and requirements.txt. See Installation below.
What you get
| Capability | Description |
|------------|-------------|
| Liquidity detection | Scans order books and pools for depth, gaps, and imbalance. |
| Hidden walls | Surfaces large buy/sell walls and their changes. |
| Multi-exchange | Built to plug into Binance, Bybit, Kraken, OKX, and others. |
| Alerts | Configurable notifications when liquidity events fire. |
| Trading framework | Modular so you can add execution, risk, or dashboards. |
Use it for liquidity grabs, order book imbalance strategies, market microstructure research, and algorithmic trading—whether you trade manually or automate.
How liquidity hunting works
Liquidity hunting targets zones where lots of orders sit (e.g. stop-loss clusters). When price sweeps those levels, liquidity is “taken” and price can move fast. This bot helps you find and watch those zones.
Find where large stop-loss clusters or thin book zones sit.
Detect liquidity sweeps and wall removals in real time.
Alert so you can enter when liquidity is taken or book vacuum appears.
Extend with your own execution (manual or automated).
Signals you can get: stop-loss clusters, sudden order book vacuum, liquidity wall removal, aggressive market order flow.
Installation
git clone https://github.com/asonglin/crypto-liquidity-ai-trading-bot.git
cd crypto-liquidity-ai-trading-bot
// App entry is app.js; configure API keys and exchanges in config.
// Core logic lives in trade/ (liquidity, orderbook) and modules/ (api, DB).
# If using a Python wrapper:
from liquidity_hunting import LiquidityBot
bot = LiquidityBot(api_key="YOUR_API_KEY", secret="YOUR_SECRET")
bot.scan_liquidity()
bot.generate_alerts()
Supported exchanges
Extensible to any exchange with a REST/WS API. Commonly used with:
What is liquidity hunting?
A strategy that focuses on levels where lots of stop-loss or passive orders sit; when those levels are hit, liquidity is consumed and price often moves sharply.
Is the bot fully automated?
It focuses on detection and alerts. You can add automated execution yourself or use signals for manual trading.
Who is it for?
Developers, quants, and algo traders who want liquidity-aware signals and a clear, extensible codebase (Node/JS, optional Python).
Contributing
We welcome pull requests and issues. Fork → branch → PR. See CONTRIBUTING.md if present.
A Claude Code plugin that automatically captures everything Claude does during your coding sessions, compresses it with AI (using Claude's agent-sdk), and injects relevant context back into future sessions.
