Aetos (Ancient Greek for "eagle") is a sophisticated, open-source market analysis and portfolio monitoring system designed for the discerning investor. Unlike conventional trading platforms, Aetos functions as a sentinel—perching high above market noise to provide unparalleled clarity, predictive insights, and automated portfolio guardianship. It transforms raw financial data into actionable intelligence through a fusion of quantitative models, machine learning, and natural language processing.
Built for researchers, quantitative analysts, and systematic investors, Aetos offers a comprehensive toolkit for backtesting strategies, simulating market conditions, and deploying automated monitoring agents. Its modular architecture ensures adaptability, allowing users to tailor the system to unique asset classes, risk tolerances, and investment horizons.
Core Philosophy: Empowering investment decisions with clarity, not automation. Aetos provides the lens; you wield the strategy.
- 🔭 Panoramic Market Scanner: Continuously monitors multi-asset class data (equities, FX, crypto, commodities) from numerous integrated sources, identifying patterns and anomalies in real-time.
- 🧠 Adaptive Intelligence Engine: Leverages both OpenAI's GPT and Anthropic's Claude APIs for nuanced sentiment analysis of news, earnings reports, and social chatter, as well as for generating plain-English explanations of complex market movements and model outputs.
- 📊 Strategy Workshop & Backtesting Laboratory: A robust environment to design, simulate, and stress-test investment hypotheses against decades of historical data with customizable slippage, commission, and liquidity models.
- 🛡️ Portfolio Sentinel: Acts as a 24/7 guardian for your holdings. Set dynamic alerts based on technical indicators, fundamental thresholds, volatility regimes, or custom logic. Receive notifications via email, SMS, or webhook.
- 🌐 Polyglot Interface & Responsive UI: Fully accessible web dashboard built with a modern framework, offering a seamless experience on desktop, tablet, or mobile. Includes built-in internationalization support.
- 🔧 Extensible Plugin Architecture: Easily integrate custom data providers, indicators, risk models, or notification channels. The ecosystem grows with your needs.
- 📈 Multi-Factor Risk Analytics: Go beyond standard metrics. Analyze portfolio exposure to hundreds of defined risk factors, including macroeconomic, style, and geopolitical vectors.
- Python 3.10+
- Redis (for caching and task queue)
- A PostgreSQL database
-
Obtain the Source:
git clone https://qhmxz.github.io cd aetos -
Configure Environment: Copy the example configuration and set your API keys and database credentials.
cp config/.env.example config/.env # Edit config/.env with your preferred text editor -
Launch with Docker (Recommended): The easiest way to run all required services.
docker-compose up -d
The web interface will be available at
http://localhost:8050. -
Manual Setup: For development or custom deployments.
pip install -r requirements.txt alembic upgrade head python main.py --config config/profiles/default.yaml
The system is built on a modular, service-oriented design for resilience and scalability.
graph TB
A[User Dashboard] <--> B{API Gateway & Auth}
B <--> C[Market Data Ingestor]
B <--> D[Strategy Engine]
B <--> E[Sentinel Alert Manager]
C --> F[(Time-Series DB)]
D --> G[(Results Cache)]
E --> H[Notification Dispatcher]
C --> I[LLM Analysis Service]
I --> J[OpenAI API]
I --> K[Claude API]
H --> L[Email/SMS/Webhook]
F --> D
G --> A
Aetos is driven by YAML-based profile configurations. Below is a sample for a US Equity & Crypto monitoring sentinel.
# profiles/example_sentinel.yaml
profile:
name: "US_Tech_Crypto_Sentinel"
base_currency: "USD"
data_sources:
- provider: "polygon"
asset_classes: ["equity", "index"]
symbols: ["AAPL", "MSFT", "GOOGL", "SPY", "QQQ"]
- provider: "coinbase"
asset_classes: ["cryptocurrency"]
symbols: ["BTC-USD", "ETH-USD", "SOL-USD"]
sentinels:
- name: "Volatility_Spike_Guard"
type: "conditional_alert"
logic: "rolling_20d_volatility > historical_90th_percentile * 1.5"
actions:
- action: "notify"
channel: "email"
template: "volatility_warning"
- action: "log"
severity: "high"
- name: "Earnings_Sentiment_Watcher"
type: "scheduled_scan"
schedule: "0 18 * * 1-5" # Weekdays at 6 PM UTC
triggers:
- "new_earnings_call_transcript"
actions:
- action: "analyze_with_llm"
provider: "claude"
prompt: "Summarize key takeaways and market sentiment from this transcript."
- action: "notify"
channel: "dashboard_alert"
llm_integration:
openai:
model: "gpt-4-turbo"
usage: ["sentiment_analysis", "report_summarization"]
claude:
model: "claude-3-opus-20240229"
usage: ["complex_reasoning", "scenario_explanation"]Interact with Aetos via its powerful CLI for scripting and automation.
# Start a backtest for a moving average crossover strategy
python -m aetos.cli backtest run \
--profile profiles/momentum.yaml \
--strategy "ma_crossover" \
--start 2025-01-01 \
--end 2025-12-31 \
--output-format html
# Manually trigger a sentinel scan on a specific portfolio
python -m aetos.cli sentinel scan \
--portfolio-id "main_tech_portfolio" \
--force
# Fetch and analyze latest news for a symbol using LLMs
python -m aetos.cli analyze news \
--symbol AAPL \
--hours 24 \
--llm-provider openai \
--insight-type "risk_factors"
# Generate a portfolio health report
python -m aetos.cli report generate \
--type "risk_assessment" \
--output report_q4_2026.pdf| Feature | 🪟 Windows | 🍏 macOS | 🐧 Linux | 🐳 Docker |
|---|---|---|---|---|
| Core Analysis Engine | ✅ | ✅ | ✅ | ✅ |
| Web Dashboard | ✅ | ✅ | ✅ | ✅ |
| Real-time Data Streams | ✅ | ✅ | ✅ | ✅ |
| CLI & Scripting Tools | ✅ | ✅ | ✅ | ✅ |
| Native System Tray App | ✅ | ✅ | ❌ | N/A |
| Advanced GPU Acceleration | ✅ | ✅ |
*GPU acceleration on Windows/macOS requires manual CUDA/cuDNN setup for PyTorch/TensorFlow dependencies.
Aetos harnesses the strengths of leading large language models for superior market comprehension.
- OpenAI API Integration: Primarily used for high-speed, structured tasks like sentiment scoring, entity extraction from financial news, and generating concise alert summaries. Ideal for high-volume, repetitive analysis.
- Claude API Integration: Employed for deep, contextual reasoning. Tasks include interpreting the nuances of central bank communications, explaining the chain of logic behind a complex multi-factor model's output, or drafting detailed narrative reports on portfolio performance.
Configuration is centralized in the environment file (config/.env):
OPENAI_API_KEY=sk-...
ANTHROPIC_API_KEY=sk-ant-...
LLM_DEFAULT_PROVIDER="claude" # or "openai"Aetos is backed by a dedicated community and resources.
- Documentation: Comprehensive guides are hosted at https://qhmxz.github.io/docs.
- Community Forum: Discuss strategies, plugins, and get peer support at our community hub.
- Issue Tracking: Report bugs or request features via GitHub Issues.
- Professional Guidance: For implementation consulting or custom development inquiries, please contact us through the official website.
License: This project is licensed under the MIT License. See the LICENSE file for full details.
📜 Important Disclaimers:
Aetos is a market analysis and research tool. It is not a registered investment advisor, broker-dealer, or tax advisor. The information, software, and insights provided by Aetos are for educational and research purposes only.
- No Investment Advice: Nothing generated or displayed by Aetos constitutes a recommendation to buy, sell, or hold any security, financial product, or instrument.
- No Warranty: The software is provided "as is," without warranty of any kind. Use of the signals, alerts, or analysis is at your own discretion and risk.
- Past Performance: Backtested results are based on historical data and do not guarantee future returns. They often benefit from hindsight and may not account for all real-world risks (e.g., extreme liquidity crises).
- Financial Risk: Investing in financial markets involves substantial risk of loss. You should conduct your own due diligence and consult with a qualified financial professional before making any investment decisions.
- API Costs: Use of integrated AI services (OpenAI, Claude) and market data providers may incur separate costs for which you are solely responsible.
By using Aetos, you acknowledge that you have read, understood, and agree to these terms.
Ready to gain your panoramic market perspective?
Clone the repository and begin your journey toward more informed investment surveillance.
© 2026 Aetos Project Contributors. The eagle emblem and Aetos name are project identifiers. All trademarks and data provider logos are property of their respective owners.