Skip to content

README.md

Latest commit

 

History

History
83 lines (65 loc) · 4.98 KB

README.md

File metadata and controls

83 lines (65 loc) · 4.98 KB
Raw

AlphaSeeker API 🚀 - Your AI-Powered Crypto Market Scanner & Analysis Engine

Unlock potential trading opportunities in the fast-paced crypto markets with the AlphaSeeker API! This powerful FastAPI backend leverages technical analysis, statistical modeling, and cutting-edge AI (GPT-4o) to scan Binance Futures, evaluate signals, and identify potential setups based on your custom criteria.

Version: 1.4.1 (Corrected technical trigger logic, disabled LSTM)

⚡️ Go Beyond Simple Alerts - Get AI-Evaluated Insights! ⚡️

⚠️ IMPORTANT WARNING / DISCLAIMER:

  • EXPERIMENTAL TOOL: This software is for educational and experimental purposes ONLY. It's a tool to explore strategy ideas, NOT a guaranteed profit machine.
  • HIGH RISK: Cryptocurrency trading involves substantial risk of loss and is not suitable for every investor. This is NOT financial advice.
  • BACKTEST LIMITATIONS: Included backtests are based on simplified proxy triggers (RSI) and DO NOT reflect real-world costs or guarantee future results.
  • AI IS NOT INFALLIBLE: GPT evaluations depend on prompt quality and data provided. They can be imperfect or wrong. Critical thinking is required.
  • USE AT YOUR OWN RISK. The author is not responsible for any financial losses. Always conduct thorough Due Diligence and Your Own Research (DYOR).

🔥 Key Features

  • Blazing Fast Scans: Asynchronously analyze hundreds of Binance USDT perpetual futures using CCXT.
  • Comprehensive TA: Calculates a wide array of indicators (RSI, SMAs, EMAs, MACD, Bollinger Bands, ATR, ADX, & more).
  • Volatility Insights: Utilizes GARCH(1,1) and VaR for risk context.
  • Intelligent Signal Triggering: Combines indicators (currently RSI + SMA Trend + ADX) to pinpoint initial potential setups.
  • 🤖 AI-Powered Evaluation (GPT-4o Ready): Sends triggered signals to OpenAI's GPT for an in-depth evaluation against the full market picture:
    • Confirms or rejects initial signals based on confluence or contradictions.
    • Suggests tactical entry, stop-loss, and take-profit levels.
    • Provides an AI confidence score for evaluated setups.
  • Proxy Backtesting: Get a quick historical perspective with a built-in (simplified RSI-based) backtester.
  • Deeply Customizable Filtering: Filter scan results by GPT confidence, backtest score, R/R ratio, trade count, win rate, profit factor, ADX, SMA alignment, and more!
  • Clean FastAPI Interface: Easy to integrate and interact with via a REST API.

🛠️ Installation & Quick Start

  1. Prerequisites: Python 3.8+, OpenAI API Key.
  2. Clone: 
    git clone <your_github_repo_url>
cd AlphaSeeker-API
  3. Setup Environment: 
    python -m venv venv
# Windows: .\venv\Scripts\activate | macOS/Linux: source venv/bin/activate
pip install -r requirements.txt
cp .env.example .env
  4. Add API Key: Edit the .env file and paste your OpenAI key: 
    OPENAI_API_KEY="your_real_openai_api_key_here"
    (Remember: Keep your .env file secure and out of Git!)
  5. Run the Beast! (Make sure your script is named main_api.py) 
    uvicorn main_api:app --host 0.0.0.0 --port 8029 --reload
  6. Explore: Open your browser to http://127.0.0.1:8029/docs for the interactive API docs (Swagger UI).

⚙️ API Endpoints

  • GET /api/crypto/tickers: Get the list of scannable Binance USDT-M perpetuals.
  • POST /api/crypto/analyze: Deep-dive analysis on a single symbol.
  • POST /api/crypto/scan: Unleash the market scanner! Customize filters in the request body (ScanRequest) to find opportunities matching your criteria.

Understanding the Workflow

  1. Data Fetch & TA: Gets historical data and calculates all indicators.
  2. Technical Trigger: Checks if the current state meets predefined technical conditions (e.g., RSI + Trend + ADX).
  3. AI Evaluation (Conditional): If a technical trigger fires, it asks GPT to evaluate the signal's strength, considering all indicators, and suggest trade parameters.
  4. Proxy Backtest: Runs a simple RSI-based backtest for historical context (independent of GPT's evaluation).
  5. Filtering & Ranking: Applies your filters (ScanRequest) to the results and ranks the survivors.

Limitations & The Road Ahead

AlphaSeeker is a powerful starting point, but remember:

  • Finding consistent "alpha" is the holy grail - this tool aids the search but doesn't guarantee discovery.
  • The default technical trigger is just one example; experimentation is encouraged!
  • GPT's effectiveness is tied to prompt engineering and the quality of the input data.
  • Backtest scores are simplified proxies.

Future Ideas: More trigger strategies, enhanced backtesting (fees, slippage, drawdown), more sophisticated filtering, alternative AI models.

Contributing

Got ideas? Found a bug? Contributions are welcome! Please open an issue or submit a Pull Request.