SEC MCP

# Research Administrator MCP ## Overview The Research Administrator is a master orchestrator MCP that coordinates all 7 financial data MCPs to generate comprehensive investment research reports. It can produce professional-grade analysis in 15 minutes that would normally take 5+ hours of manual research. ## Features - **Intelligent Orchestration**: Coordinates multiple MCPs in parallel for efficient data collection - **Multiple Research Depths**: From 5-minute quick checks to 2-hour deep dives - **Natural Language Processing**: Understands research requests in plain English - **Professional Templates**: Generates reports formatted like top-tier Wall Street research - **Comprehensive Coverage**: Combines SEC filings, sentiment, analyst ratings, alternative data, and more ## Research Depth Levels ### 1. Quick Check (5 minutes) - Current valuation vs peers - Basic DCF analysis - Analyst consensus - Simple Buy/Hold/Sell recommendation ### 2. Standard Analysis (15 minutes) - DEFAULT - Executive summary with key findings - DCF model with sensitivity analysis - Multi-source sentiment analysis - Institutional flow assessment - Alternative data signals - Bull/bear cases with evidence - Risk factor analysis - 12-month price target ### 3. Deep Dive (45 minutes) - Comprehensive 10-15 page research report - Multiple DCF scenarios - Historical performance patterns - Competitive positioning - Management assessment - Detailed risk/reward framework ### 4. Sector Analysis (2+ hours) - Entire sector overview - Top 10 companies ranked - Industry-wide DCF comparison - Relative value analysis - Thematic opportunities ## Available Tools ### create_research_plan Creates a comprehensive research plan for a stock. **Example:** ```json { "ticker": "AAPL", "depth": "standard", "sector": "technology" } ``` ### parse_research_request Converts natural language into MCP commands. **Example:** ```json { "request": "Generate a quick investment check for TSLA" } ``` ### generate_report_outline Creates a report outline based on collected data. ### list_available_mcps Shows all available MCPs and their tools. ### generate_executive_summary Creates an executive summary from research findings. ## How It Works The Research Administrator coordinates these MCPs: 1. **SEC_SCRAPER**: Financial statements, SEC filings 2. **NEWS_SENTIMENT**: News sentiment from multiple sources 3. **ANALYST_RATINGS**: Wall Street ratings and targets 4. **INSTITUTIONAL**: Smart money flows, insider trading 5. **ALTERNATIVE_DATA**: Job postings, web traffic, patents 6. **INDUSTRY_ASSUMPTIONS**: DCF parameters, sector metrics 7. **ECONOMIC_DATA**: Macro indicators, Fed data ## Example Usage in Claude Desktop ### Quick Investment Check ``` "Use RESEARCH_ADMINISTRATOR to create a quick investment check for NVDA" ``` ### Standard Analysis ``` "Use RESEARCH_ADMINISTRATOR to perform comprehensive analysis for MSFT" ``` ### Deep Research ``` "Use RESEARCH_ADMINISTRATOR to conduct deep research on AMZN with focus on cloud growth" ``` ### Natural Language Request ``` "Use RESEARCH_ADMINISTRATOR to analyze whether Apple is a good buy right now" ``` ## Integration with Claude Desktop 1. Add to your `claude_desktop_config.json`: ```json { "RESEARCH-ADMIN": { "command": "/Users/LuisRincon/SEC-MCP/FinancialMCPs/RESEARCH_ADMINISTRATOR/start-mcp.sh" } } ``` 2. Restart Claude Desktop 3. The Research Administrator will coordinate all other MCPs automatically ## Research Workflow When you request research, the Administrator: 1. **Parses your request** to understand ticker, depth, and focus areas 2. **Creates a research plan** with specific MCP calls 3. **Orchestrates data collection** in optimal parallel/sequential order 4. **Provides execution instructions** for Claude to follow 5. **Generates report templates** for consistent formatting ## Output Format Research reports include: - **Executive Summary**: Key findings and recommendation - **Financial Overview**: Valuation metrics and comparisons - **DCF Analysis**: Fair value with sensitivity - **Sentiment Analysis**: News, social, and analyst sentiment - **Institutional Activity**: Smart money flows - **Alternative Signals**: Hiring, web traffic, patents - **Investment Thesis**: Bull and bear cases - **Risk Assessment**: Key risks and mitigants - **Price Target**: 12-month target with rationale ## Best Practices 1. **Start with Standard Analysis** (15 min) for most stocks 2. **Use Quick Check** for rapid screening of multiple stocks 3. **Reserve Deep Dive** for high-conviction ideas 4. **Specify sector** for more accurate DCF assumptions 5. **Add focus areas** for targeted analysis (e.g., "focus on AI revenue") ## Limitations - Relies on web scraping (respects robots.txt) - Data freshness depends on source update frequency - Some alternative data may be estimates - Not suitable for real-time trading ## Example Research Commands ### Technology Stock Analysis ``` { "ticker": "GOOGL", "depth": "standard", "sector": "technology", "focus_areas": ["AI monetization", "cloud growth"] } ``` ### Financial Sector Deep Dive ``` { "ticker": "JPM", "depth": "deep", "sector": "finance", "focus_areas": ["interest rate sensitivity", "credit quality"] } ``` ### Quick Biotech Check ``` { "ticker": "MRNA", "depth": "quick", "sector": "healthcare" } ``` ## Troubleshooting If the Research Administrator has issues: 1. Ensure all 7 financial MCPs are properly connected 2. Check that you have internet connectivity 3. Verify the virtual environment is activated 4. Review logs for specific MCP errors ## Future Enhancements - Real-time monitoring mode - Portfolio-level analysis - Custom report templates - Excel/PDF export - Comparative analysis across multiple tickers - ESG integration