Summary

MaverickMCP is a locally-run FastMCP 2.0 server designed for individual traders and investors, offering professional-grade financial data analysis, technical indicators, and portfolio optimization tools. It integrates seamlessly with AI interfaces like Claude Desktop, providing comprehensive stock analysis capabilities without authentication or billing complexity, thereby democratizing access to advanced financial tools.

How It Works

This project leverages a modern Python stack with uv for dependency management and FastAPI for its web server. It employs a FastMCP 2.0 architecture, supporting HTTP, SSE, and STDIO transports for flexible client integration. Core functionalities include a pre-seeded S&P 500 database, a VectorBT-powered backtesting engine, and advanced AI research agents orchestrated via LangGraph. Redis powers a smart caching layer for performance, with fallbacks to in-memory storage, ensuring efficient data retrieval and analysis.

Quick Start & Requirements

• Primary Install/Run: Clone the repo, then uv sync followed by make dev to install dependencies and start the server.
• Prerequisites: Python 3.12+, uv (recommended), TA-Lib (essential for technical analysis, installation varies by OS), Redis (optional, for enhanced caching), PostgreSQL or SQLite (optional, for data persistence).
• API Keys: Tiingo API key (required for stock data), OpenRouter/OpenAI/Anthropic API keys (recommended for AI research), Exa/Tavily API keys (recommended for web search).
• Links: Tiingo: https://www.tiingo.com/, TA-Lib installation guides provided within the README.

Highlighted Details

• Pre-seeded Data: Includes all 520 S&P 500 stocks with comprehensive screening recommendations.
• Advanced Backtesting: VectorBT-powered engine with 15+ built-in strategies, ML algorithms, walk-forward optimization, and Monte Carlo simulations.
• AI Research Agents: Orchestrated via LangGraph for deep company and market analysis, offering 7-256x speedups, cost optimization, and content filtering.
• Portfolio Management: Features personal portfolio tracking with cost basis averaging, live P&L, and correlation analysis.
• Integration: Native MCP support for Claude Desktop, Cursor IDE, and other clients via HTTP, SSE, and STDIO transports.

Maintenance & Community

The project mentions "community-driven development" but provides no specific links to community channels (e.g., Discord, Slack), active contributors, or a public roadmap.

Licensing & Compatibility

• License: MIT License.
• Compatibility: Permissive for personal and commercial use, allowing integration into closed-source applications.

Limitations & Caveats

The TA-Lib installation can be complex. Full functionality, especially for AI research and real-time data, requires obtaining and configuring multiple third-party API keys. The project carries a strong disclaimer stating it is for educational purposes only and not financial advice, with users assuming all investment risks.