--- title: Financial Research Agent emoji: 📊 colorFrom: blue colorTo: green sdk: gradio sdk_version: 4.11.0 app_file: app.py pinned: false license: mit --- # 📊 Financial Research Agent **Multi-agent equity analysis combining SEC filings with real-time market intelligence** ## 🎯 What This Does Analyzes stocks using a multi-agent AI system that: - **SEC Filing Agent**: Analyzes 10-K/10-Q filings with FinBERT + SEC-BERT - **Market Intelligence Agent**: Gathers real-time price data, technicals (RSI/MACD), and news - **Synthesis Agent**: Cross-references fundamentals vs. market action for final recommendation ## 🚀 How to Use 1. Enter a stock ticker (e.g., TSLA, AAPL, RBLX) 2. Optionally add company name for better news search 3. Select SEC filing type (10-K for annual, 10-Q for quarterly) 4. Click **Analyze** 5. Wait ~1-2 minutes for comprehensive analysis ## 🔬 What Makes This Different Unlike basic sentiment tools, this platform: - ✅ Analyzes **actual SEC filings** (not just news) - ✅ Uses **LIME explainability** to show which words drive decisions - ✅ **Cross-validates** what companies say (filings) vs. what markets do (price) - ✅ Provides **evidence-based** recommendations with risk factors ## ⚙️ Configuration **Required Environment Variables:** - `SEC_EMAIL`: Your email (SEC EDGAR compliance) - `NEWS_API_KEY`: NewsAPI key ([get free tier here](https://newsapi.org/)) **Note**: First run downloads ML models (~500MB) - subsequent analyses are much faster! ## 📊 Example Output ``` Market Sentiment: BULLISH Confidence: HIGH Key Risks: • [HIGH] Increasing competition from major gaming platforms • [MEDIUM] User acquisition costs trending upward Recommended Action: BUY - Favorable risk/reward ``` ## 🛠️ Tech Stack - **NLP Models**: FinBERT, SEC-BERT - **Explainability**: LIME - **Data Sources**: SEC EDGAR, yfinance, NewsAPI - **Framework**: Multi-agent orchestration - **Interface**: Gradio ## 📝 Limitations - SEC filings updated quarterly/annually (not real-time) - NewsAPI free tier: 100 requests/day - First analysis takes longer (model download) - Analysis time: 1-3 minutes depending on data availability ## 🔗 Links - **GitHub**: [SanchitSharma10/financial-research-agent](https://github.com/SanchitSharma10/financial-research-agent) - **Author**: [Sanchit Sharma](https://linkedin.com/in/sanchit-sharma10) ## ⚠️ Disclaimer This tool is for **research and educational purposes only**. Not financial advice. Always conduct your own due diligence before making investment decisions. ## 📄 License MIT License - See [LICENSE](LICENSE) file