AIBrokers

AIBrokers

The first real-world AI hedge fund framework in crypto, fully open source! AIBrokers is a framework designed to create multi AI agents for managing hedge funds. These AI agents act as professional traders, operating 24/7 to manage investments for their owners. My vision and aspiration are that, with unlimited contributions from the community, AIBrokers can outperform hedge funds, whales, and market makers in the future. AIBrokers is open source and welcomes contributions from anyone.

This system employs several agents working together:

• Trader Behavior Agent: Gather on-chain trader behavior, traders actions, etc.
• Quant Agent: calculates signals like MACD, RSI, Bollinger Bands, etc.
• Sentiment Agent: gathers, analyzes crypto market sentiment from social media, news, and on-chain data to support trading strategies, etc.
• Fundamental Agent: evaluates crypto projects’ tokenomics, on-chain data, market performance, and ecosystem to guide long-term investment decisions, etc.
• Technical Analyst Agent: analyzes crypto price charts, trends, and indicators to identify trading opportunities and optimize entry/exit points, etc.
• Risk manager: assesses market volatility, portfolio exposure, and potential risks to minimize losses and optimize risk-reward ratios.
• Fund Manager Agent: makes final trading decisions and generates orders, etc.

Note: the system simulates trading decisions, it does not actually trade.

Prerequisites

• Python 3.9 or higher (Python 2.x is not supported)
• OpenAI API key with access to GPT-4
• Poetry package manager

Setting up Python 3.9+

macOS:

# Using Homebrew
brew install [email protected]

# Verify installation
python3.9 --version

Ubuntu/Debian:

# Add deadsnakes PPA
sudo add-apt-repository ppa:deadsnakes/ppa
sudo apt update

# Install Python 3.9
sudo apt install python3.9 python3.9-venv

# Verify installation
python3.9 --version

Windows:

1. Download Python 3.9+ from python.org
2. Run the installer
   • ✅ Check “Add Python to PATH”
   • ✅ Check “Install pip”
3. Verify installation:

python --version

Switching Python Versions

If you have multiple Python versions installed:

macOS/Linux:

# List available Python versions
ls /usr/local/bin/python*

# Use specific version with Poetry
poetry env use python3.9

Windows:

# List installed Python versions
py --list

# Use specific version with Poetry
poetry env use py3.9

Setup

First, check if you have a compatible Python version:

python3 --version  # Should be 3.9 or higher

Then check if Poetry is already installed:

poetry --version

If you see a version number (like Poetry (version 2.0.1)), you can skip to step 4. If you get a “command not found” error, follow steps 1-3.

Clone the repository:

git clone https://github.com/AI-Brokers/AIBrokers.git
cd AIBrokers

1. Install Poetry:

macOS / Linux:

curl -sSL https://install.python-poetry.org | python3 -
# Or you can do that to skip step 1 and 2
# pip install -r requirements.txt

Windows (PowerShell):

(Invoke-WebRequest -Uri https://install.python-poetry.org -UseBasicParsing).Content | py -

2. Add Poetry to your PATH:

macOS / Linux:

export PATH="$HOME/.local/bin:$PATH"

Windows:

$Env:Path += ";C:\Users\$Env:USERNAME\AppData\Roaming\Python\Scripts"

3. Verify Poetry installation:

poetry --version

If you don’t see a version number, try closing and reopening your terminal.

4. Install dependencies:

cd AIBrokers
poetry install

5. Set up your environment variables:

# Create .env file for your API keys
cp .env.example .env
# Edit the .env file with your API keys using your preferred editor

Common Issues

Python Version Error

If you see an error about Python version compatibility, make sure you’re using Python 3.9 or higher:

python3 --version

Poetry Environment Issues

If you encounter Poetry-related errors:

# Create a new virtual environment
poetry env remove --all
poetry install

# Or specify Python version explicitly
poetry env use python3.11  # or your Python 3.9+ version

Usage

Running the AIBrokers

poetry run python src/main.py --crypto BTC
# Or python src/main.py --crypto BTC

Example Output:

You can also specify a --show-reasoning flag to print the reasoning of each agent to the console.

poetry run python src/main.py --crypto BTC --show-reasoning
# Or python src/main.py --crypto BTC --show-reasoning

You can optionally specify the start and end dates to make decisions for a specific time period.

poetry run python src/main.py --crypto BTC --start-date 2024-01-01 --end-date 2024-03-01 
# Or python src/main.py --crypto BTC --start-date 2024-01-01 --end-date 2024-03-01

You can customize your balances, leverage and risk for each trade for your portfolio

poetry run python src/main.py --crypto BTC --balance 500000 --leverage 20 --risk 0.01
# Or python src/main.py --crypto BTC --balance 500000 --leverage 20 --risk 0.01

Risk for each trade here is the ratio of total fund that can be lost for each trade. Example: Balance 500000 , Risk = 0.01 , that means the max loss for each trade is 5000

Running the Backtester

poetry run python src/backtester.py --crypto BTC
# Or python src/backtester.py --crypto BTC

Example Output:

Starting backtest...
Date         Crypto Action Quantity    Price         Cash    colatteralLong collateralShort  Total Value
-----------------------------------------------------------------------------------------------------------
2024-12-16   BTC    long     100000 103124.30         0.00     0.97             0              100030.57
2024-12-17   BTC    short    103816 107026.90         0.09        0             0.97           103816.19
2024-12-18   BTC    long     103397 107458.50         0.53     0.96             0              103160.69
2024-12-19   BTC    long     100433 104617.30         0.14     0.96             0              100432.75
2024-12-20   BTC    short     94863 98815.90          0.41        0             0.96            94863.67

You can optionally specify the start and end dates to backtest over a specific time period.

poetry run python src/backtester.py --crypto BTC --start-date 2024-01-01 --end-date 2024-03-01
# Or python src/backtester.py --crypto BTC --start-date 2024-01-01 --end-date 2024-03-01

Configuration

Analysis Weights

You can customize the weights of different analysis components in the portfolio management decision-making process. These weights are defined in src/config/analysis_weights.py:

• TECHNICAL_ANALYSIS_WEIGHT: Weight given to technical analysis (default: 25%)
• SENTIMENT_ANALYSIS_WEIGHT: Weight given to sentiment analysis (default: 10%)

To modify these weights, simply update the values in the configuration file according to your trading strategy preferences.

Project Structure

AIBrokers/
├── src/
│   ├── agents/                   # Agent definitions and workflow│
│   │   ├── market_data.py        # Market data agent
│   │   ├── portfolio_manager.py  # Portfolio management agent
│   │   ├── risk_manager.py       # Risk management agent
│   │   ├── sentiment.py          # Sentiment analysis agent
│   │   ├── state.py              # Agent state
│   │   ├── technicals.py         # Technical analysis agent│
│   ├── tools/                    # Agent tools
│   │   ├── api.py                # API tools
│   ├── backtester.py             # Backtesting tools
│   ├── main.py # Main entry point
├── pyproject.toml
├── ...

Contributing

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Create a Pull Request

Usercase

• Chat Box/ Social Bot/ Signal Bot
• The Autonomous AI-Driven Hedge Fund
• AI-Powered Copilot Trading Assistant

Challenges/Task

Here’s a structured to-do list to request contributions from the community for the AIBrokers open-source project. It includes both challenges and tasks to inspire developers to contribute meaningfully:

1. Core System Architecture

• Define and document APIs for communication between agents (e.g., Trader Behavior Agent ↔ Fund Manager Agent).

• Implement agent orchestration: Develop a modular framework to coordinate multiple agents effectively.

• Set up a backtesting environment for simulated trading strategies to validate AI decisions.

2. Trader Behavior Agent

• Build a scraper to gather on-chain trader behavior data (e.g., wallet actions, trading volume).

• Implement anomaly detection for unusual trading patterns.

• Develop tools for visualizing trader behavior trends.

3. Quant Agent

• Implement advanced indicators (e.g., MACD, RSI, Bollinger Bands, Ichimoku Cloud).

• Create a signal combination engine for multi-indicator strategies.

• Integrate with live and historical price data APIs (e.g., Binance, Kraken).

4. Sentiment Agent

• Develop a social media sentiment analyzer using NLP (Twitter, Reddit, Telegram, etc.).

• Incorporate on-chain data for community sentiment trends (e.g., token transfers).

• Build a dashboard for sentiment visualization with real-time updates.

5. Fundamental Agent

• Design a scoring model for evaluating tokenomics (e.g., inflation rate, supply schedule).

• Analyze crypto ecosystems and detect growth opportunities.

• Build tools for tracking developer activity on GitHub and other repositories.

6. Technical Analyst Agent

• Develop AI models to predict price trends based on historical chart patterns.

• Implement multi-timeframe analysis (e.g., 15m, 1H, daily charts).

• Automate entry and exit signal generation based on technical indicators.

7. Risk Manager

• Build a module to calculate portfolio exposure and suggest optimal rebalancing.

• Develop algorithms for stop-loss and take-profit management.

• Implement a stress testing tool for simulating market crashes.

8. Fund Manager Agent

• Create a decision-making engine to rank and execute agent recommendations.

• Develop an order simulation system (e.g., buy/sell orders with slippage considerations).

• Allow customizable strategies: e.g., risk tolerance, aggressive vs. conservative.

9. Infrastructure & Scalability

• Migrate data pipelines to a scalable distributed system (e.g., Kafka, RabbitMQ).

• Ensure agent modularity to support plug-and-play integration.

• Containerize the framework using Docker for easy deployment.

10. Community Contributions

• Build detailed contributing guidelines (e.g., coding standards, testing requirements).

• Open a list of Good First Issues for new contributors.

• Encourage building plugins or extensions for specific strategies.

11. Challenges for the Community

• How can AI agents detect manipulated or wash-traded tokens?

• What’s the best way to optimize multi-agent collaboration for decision-making?

• Can reinforcement learning improve trading strategy performance over time?

• How do we handle low-liquidity markets effectively?

12. Documentation and Knowledge Sharing

• Write clear documentation for each agent, its role, and integration process.

• Publish examples of successful backtests to showcase the system’s potential.

• Create video tutorials to onboard developers.

13. Data Brokers

By addressing these areas, the project can attract developers of varying skill levels to contribute and grow AIBrokers into a comprehensive framework.

Community

• X: https://x.com/aibrokers_xyz
• Discord: https://discord.gg/zQpKw6eeQu
• Media Kit: https://drive.google.com/drive/folders/1SKjgkHd0j-iClgCxsuVcYfHyxB-cvqhI?usp=sharing

---

• Schema Inspired: https://github.com/virattt/ai-hedge-fund

License

This project is licensed under the MIT License - see the LICENSE file for details.