AI-powered research platform for crypto and traditional finance with Web UI + optional Notion automation

A production-ready research and market intelligence platform powered by OpenRouter (50+ AI models). Analyze cryptocurrencies and stocks side-by-side, run comprehensive financial research, and automate due diligence workflows.

1. Open the Streamlit web interface
2. Upload documents, add URLs, or enter a research query
3. Use Financial Research for stocks or Crypto AI for crypto
4. Compare assets across markets with Market Intelligence
5. Get comprehensive research reports instantly

1. Connect your Notion database
2. Add a project to Notion
3. Agent automatically detects the new entry
4. AI researches, scores, and evaluates
5. Full report published back to Notion

• Price Data: Real-time snapshots and historical prices via OpenBB
• Fundamentals: Income statements, balance sheets, cash flow statements
• SEC Filings: 10-K, 10-Q, 8-K with item-level extraction
• Metrics: Financial ratios, analyst estimates, insider trades
• News: Company news and market sentiment
• AI Router: Natural language queries routed to appropriate tools

• Live Market Data: Real-time cryptocurrency prices via CoinGecko MCP
• Historical Analysis: Price history with customizable intervals
• Top Coins: Access to top 250 cryptocurrencies by market cap
• AI Chat: Interactive crypto analysis and insights
• Portfolio Tracking: Multi-coin comparisons and trends

• Direct Comparison: Compare crypto vs stocks (BTC vs V, ETH vs MA)
• Volatility Analysis: Normalized comparisons accounting for 24/7 vs market hours
• Sector Views: CMC-style sector categorization for tech stocks
• Yield Comparison: Staking APY vs dividend yields

• Multi-Format Documents: PDF, DOCX, TXT, Markdown with OCR
• DocSend Integration: Automated presentation analysis
• Web Scraping: Firecrawl-powered sitemap discovery and extraction
• Deep Research: LangChain ODR framework for multi-agent research
• RAG Chat: Context-aware Q&A using FAISS vector search

• Structured Extraction: People, organizations, funding rounds, metrics
• Source Grounding: All entities linked to source documents
• Smart Caching: Results cached for unchanged content

• Real-Time Monitoring: Watches Notion database for new entries
• Auto-Research: Triggers deep research on new projects
• AI Scoring: Automated due diligence evaluation
• Direct Publishing: Reports published to Notion pages

├── main.py                          # Streamlit entry point
├── src/
│   ├── controllers/
│   │   └── app_controller.py        # Auth, routing, validation
│   ├── pages/
│   │   ├── interactive_research.py  # Document processing & AI analysis
│   │   ├── financial_research.py    # Stocks research UI
│   │   ├── market_intelligence.py   # Cross-asset comparison
│   │   ├── crypto_chatbot.py        # Crypto AI interface
│   │   └── notion_automation.py     # Notion CRM integration
│   ├── services/
│   │   ├── financial_tools/         # 19 financial data tools
│   │   │   ├── tools.py             # Tool implementations
│   │   │   ├── router.py            # OpenRouter-based routing
│   │   │   ├── schemas.py           # OpenAI function schemas
│   │   │   └── crypto_resolver.py   # Ticker format conversion
│   │   ├── market_intelligence/     # Cross-asset services
│   │   │   ├── cross_asset_service.py
│   │   │   ├── volatility_service.py
│   │   │   └── sector_service.py
│   │   ├── openbb/                  # OpenBB Platform client
│   │   │   └── client.py
│   │   └── mcp/                     # MCP integrations
│   │       └── coingecko_client.py  # CoinGecko data
│   ├── core/
│   │   ├── research_engine.py       # Research automation
│   │   ├── scanner_utils.py         # Web discovery (Firecrawl)
│   │   └── docsend_client.py        # DocSend processing
│   └── utils/
│       └── session_persistence.py   # Secure session management
├── config/
│   ├── users.yaml                   # User management
│   ├── mcp_config.json              # MCP integrations
│   └── sector_classifications.yaml  # Sector mappings
└── tests/

• Backend: Python 3.11+, Streamlit
• Financial Data: OpenBB Platform, CoinGecko, FMP
• AI/ML: OpenRouter, LangChain, FAISS
• Browser Automation: Selenium, Playwright
• Document Processing: PyMuPDF, Tesseract OCR
• Visualization: Plotly, Altair

# Clone repository
git clone https://github.com/sheeki03/analystOS.git
cd analystOS

# Install dependencies
pip install -r requirements.txt

# Install browser dependencies
playwright install

# Install Tesseract OCR (for document processing)
# macOS: brew install tesseract
# Ubuntu: sudo apt-get install tesseract-ocr

Create a .env file:

# Required: AI Model Access
OPENROUTER_API_KEY=your_openrouter_key

# Financial Data (Stocks)
FMP_API_KEY=your_fmp_key              # financialmodelingprep.com
OPENBB_PAT=your_openbb_token          # Optional, for higher limits

# Optional: Additional Providers
FINNHUB_API_KEY=your_finnhub_key      # For news
OPENAI_API_KEY=your_openai_key
ANTHROPIC_API_KEY=your_anthropic_key

# Optional: Notion Integration
NOTION_TOKEN=your_notion_token

# Optional: Web Scraping
FIRECRAWL_API_KEY=your_firecrawl_key

# Optional: Deep Research
TAVILY_API_KEY=your_tavily_key

# Security
SESSION_SECRET_KEY=your_random_secret_key

streamlit run main.py

Visit http://localhost:8501 and login.

• USD pairs only: BTC-USD supported, BTC-ETH not supported
• No minute interval: Use day, week, month, or year
• Max 365 days: Historical data limited to 1 year from today
• Close prices only: Returns close + volume, not OHLC

The platform comes with a default cross-asset watchlist:

Crypto: BTC, ETH, SOL Payments: V (Visa), MA (Mastercard) Tech: NVDA, INTC, MSFT, GOOGL Finance: JPM, COIN, SOFI

Query: "Show me AAPL revenue for the last 3 years"
→ Routes to get_income_statements(ticker="AAPL", period="annual", limit=3)

Query: "Compare NVDA and AMD financial metrics"
→ Routes to get_financial_metrics for both tickers

Query: "Bitcoin price last month"
→ Routes to get_crypto_prices(ticker="BTC-USD", interval="day", ...)

Query: "Top 10 cryptocurrencies"
→ Uses CoinGecko markets endpoint

Query: "Compare BTC vs VISA market cap"
→ Uses Market Intelligence cross-asset service

• HMAC-Signed Sessions: URL session tokens are cryptographically signed
• Input Validation: Username/password validation with strength requirements
• Subprocess Whitelist: Only approved commands can execute
• SSL Verification: Never disabled, fails safely
• Role-Based Access: Admin vs researcher permissions
• XSS Prevention: Safe Streamlit rendering methods

Supported via OpenRouter:

# Run all tests
pytest

# Run with coverage
pytest --cov=src tests/

# Run specific tests
pytest tests/integration/test_tool_schemas.py
pytest tests/services/test_crypto_resolver.py

• Added OpenBB integration for stocks/fundamentals
• Added 19 financial data tools
• Added Market Intelligence with cross-asset comparison
• Added Financial Research and Market Intelligence pages
• Enhanced CoinGecko client with markets endpoint
• Integrated Firecrawl for sitemap scanning
• Security hardening (SSL, HMAC sessions, input validation)

• Deep Research (ODR) integration
• Enhanced crypto intelligence
• Entity extraction (LangExtract)

• Notion automation
• Crypto chatbot

• Basic research and document processing

1. Fork the repository
2. Create feature branch (git checkout -b feature/amazing-feature)
3. Commit changes (git commit -m 'Add amazing feature')
4. Push to branch (git push origin feature/amazing-feature)
5. Open Pull Request

MIT License - see LICENSE for details.

• Issues: GitHub Issues
• Discussions: GitHub Discussions

Built for researchers, analysts, and investors who need comprehensive market intelligence across crypto and traditional finance.