AI-Powered Multi-Agent Stock Analysis for Claude Code
Features • Installation • Usage • Agents • How It Works
Super Hedge Fund Skill is a Claude Code skill that transforms Claude into a multi-agent hedge fund analyst. It combines rule-based precision with Claude-powered insight to deliver comprehensive stock analysis.
User: "Analyze AAPL, MSFT, NVDA"
→ 8 AI Agents Analyze in Parallel
→ Risk-Adjusted Position Sizing
→ Comprehensive Markdown Report
⚠️ Disclaimer: Educational and research purposes only. NOT investment advice.
| Feature | Description |
|---|---|
| 🤖 8 Analysis Agents | Fundamental, Technical, Valuation, Sentiment + 4 Investor Personas |
| 📊 Rule-Based Precision | Deterministic calculations for financial metrics |
| 🧠 Claude-Powered Insight | Interpretive analysis from famous investor perspectives |
| ⚖️ Risk Management | Volatility-based position sizing and limits |
| 📝 Professional Reports | Markdown format with executive summary |
| 🔄 Multi-Source Data | Yahoo Finance integration with caching |
# Clone the repository
git clone https://github.com/StanleyChanH/super-hedge-fund-skill.git
# Copy to Claude Code skills directory
cp -r super-hedge-fund-skill/super-hedge-fund-skill ~/.claude/skills/# Create skills directory if not exists
mkdir -p ~/.claude/skills
# Download and extract
curl -L https://github.com/StanleyChanH/super-hedge-fund-skill/archive/refs/heads/main.tar.gz | tar xz
# Install
cp -r super-hedge-fund-skill-main/super-hedge-fund-skill ~/.claude/skills/Simply mention stock tickers or investment questions in Claude Code:
User: Analyze AAPL
User: Should I buy TSLA?
User: What's the signal for NVDA and GOOGL?
User: Compare MSFT and AMZN
User: Analyze AAPL with brief mode
User: Full analysis for TSLA, NVDA, META with $100k portfolio
User: What would Warren Buffett think of NVDA?
# 🏦 AI Hedge Fund Analysis Report
**Ticker**: AAPL
**Action**: 🟡 Hold
**Confidence**: 55%
## Agent Signals
| Agent | Signal | Confidence | Reasoning |
|-------|--------|------------|-----------|
| Fundamental | 🟢 Bullish | 82% | ROE 54%, Net Margin 25% |
| Technical | 🟡 Neutral | 52% | EMA trend mixed, RSI 55 |
| Valuation | 🔴 Bearish | 68% | P/E 34x above historical |
| Buffett | 🟢 Bullish | 75% | Wide moat, excellent ROE |
| Wood | 🟡 Neutral | 60% | AI potential but expensive |
| Burry | 🔴 Bearish | 70% | Overvalued, no catalyst |
| Lynch | 🟡 Neutral | 55% | PEG > 1, growth slowing |
**Consensus**: Neutral (Bullish: 2 / Bearish: 2 / Neutral: 3)| Agent | Metrics | Signal Logic |
|---|---|---|
| Fundamental | ROE, Net Margin, Debt/Equity, Revenue Growth | Score ≥ 60 → Bullish |
| Technical | EMA, RSI, MACD, Bollinger, Momentum | Weighted signal aggregation |
| Valuation | DCF, Owner Earnings, EV/EBITDA | Gap > 15% from intrinsic value |
| Sentiment | News headlines, Insider trades | Weighted positive/negative |
| Investor | Philosophy | Focus |
|---|---|---|
| 🧠 Warren Buffett | Value Investing | Moat, ROE > 15%, Margin of Safety |
| 🚀 Cathie Wood | Growth Investing | Disruptive Tech, AI, Innovation |
| 📉 Michael Burry | Contrarian Investing | Deep Value, Against Consensus |
| 📊 Peter Lynch | GARP | PEG < 1, Understandable Business |
┌─────────────┐ ┌─────────────┐ ┌─────────────┐
│ Input │────▶│ Fetch Data │────▶│ Run Agents │
│ (Tickers) │ │(Price/Fin) │ │(Rule+Claude)│
└─────────────┘ └─────────────┘ └─────────────┘
│
▼
┌─────────────┐ ┌─────────────┐ ┌─────────────┐
│ Output │◀────│ Aggregate │◀────│Risk Analysis│
│ (Report) │ │ Signals │ │(Vol/Limits) │
└─────────────┘ └─────────────┘ └─────────────┘
# Count signals from all agents
bullish_count = count(bullish signals)
bearish_count = count(bearish signals)
neutral_count = count(neutral signals)
# Determine consensus
if bullish_count > bearish_count + 2:
consensus = "BULLISH"
elif bearish_count > bullish_count + 2:
consensus = "BEARISH"
else:
consensus = "NEUTRAL"| Volatility | Risk Level | Max Position |
|---|---|---|
| < 15% | Low | 25% |
| 15-30% | Medium | 20% |
| 30-50% | High | 15% |
| > 50% | Very High | 10% |
super-hedge-fund-skill/
├── README.md # This file
├── README_CN.md # Chinese documentation
├── super-hedge-fund-skill/ # Skill directory
│ ├── SKILL.md # Main skill file
│ ├── scripts/ # Python modules
│ │ ├── analysts.py # Rule-based analysts
│ │ ├── data_fetcher.py # Data fetching
│ │ ├── investor_prompts.py # Claude investor prompts
│ │ └── report_generator.py # Report generation
│ ├── references/ # Reference docs
│ │ ├── investor-agents.md # Investor personas
│ │ └── analysis-methods.md # Analysis methods
│ └── assets/ # Output templates
│ └── report-template.md # Report template
Works with any US stock ticker available on Yahoo Finance:
- Large Cap: AAPL, MSFT, GOOGL, AMZN, NVDA, TSLA, META
- Mid/Small Cap: Most NYSE/NASDAQ listed stocks
- ETFs: SPY, QQQ, IWM, etc.
| Data Type | Source | Notes |
|---|---|---|
| Price Data | Yahoo Finance | OHLCV, 60+ days |
| Financials | Yahoo Finance | ROE, P/E, margins, debt |
| News | Yahoo Finance | Recent headlines |
# Run module tests
cd super-hedge-fund-skill/super-hedge-fund-skill/scripts
python -m analysts
python -m data_fetcher
python -m report_generatorContributions are welcome! Please feel free to submit a Pull Request.
- Fork the repository
- Create your feature branch (
git checkout -b feature/AmazingFeature) - Commit your changes (
git commit -m 'Add some AmazingFeature') - Push to the branch (
git push origin feature/AmazingFeature) - Open a Pull Request
This project is licensed under the MIT License - see the LICENSE file for details.
IMPORTANT: This software is for educational and research purposes only.
- ❌ NOT financial advice
- ❌ NOT a recommendation to buy or sell securities
- ❌ NOT a guarantee of future performance
Investing involves substantial risk. Past performance is not indicative of future results. Always do your own research and consult with a qualified financial advisor before making investment decisions.
- Inspired by AI Hedge Fund project
- Built for Claude Code
- Data provided by Yahoo Finance
Made with ❤️ by StanleyChanH