Maverick MCP

Community

Personal stock analysis MCP server with market data, technical analysis, screening, portfolio tracking, and optional backtesting and deep research.

GitHub source repository ↗

★ 652 Stars Category · Other Very popular Source revision 95c919f8ed22

44FMRS · D

Reliability

4/20

Security and permissions

8/20

Maintenance

10/20

Documentation

12/20

Setup experience

10/20

MaverickMCP offers a comprehensive suite of local stock analysis tools covering technical analysis, screening, portfolio management, and optional extras. Strengths: no API key needed for core features, rich toolset, and multi-client integration. Weaknesses: screening requires manual history fetching, and backtesting/research extras require additional install and API keys. Overall, it's a powerful and flexible tool for personal investment analysis.

Read the FMRS scoring method →

MaverickMCP is a FastMCP-based local stock analysis server designed for individual traders and investors. Core tools require no API key and get market data from yfinance. It provides 37 core tools covering market data, technical analysis, stock screening, portfolio management (with risk dashboard, watchlists, and trade journal). Optional extras add backtesting (VectorBT-powered) and deep research (LangGraph-based, bring-your-own LLM key). Supports stdio and streamable HTTP transports, integrating seamlessly with Claude Desktop, Cursor, Claude Code, and Windsurf.

Tools

market_data_get_price_history

OHLCV price history for a ticker, smart-cached.

market_data_get_price_history_batch

Price history for multiple tickers at once.

market_data_get_quote

A single quote, TTL-cached.

market_data_get_stock_fundamentals

Valuation, financials, and trading stats.

market_data_get_market_overview

Indices, sector performance, top movers, and volatility.

market_data_get_chart_links

Static external chart links for a ticker.

market_data_clear_market_cache

Clear cached quotes (mutates cache state).

technical_get_rsi_analysis

RSI reading and signal label.

technical_get_macd_analysis

MACD reading, signal label, and crossover state.

technical_get_support_resistance

Support/resistance levels.

technical_get_full_technical_analysis

Full technical analysis: trend, outlook, every indicator.

screening_get_bullish

Top Maverick bullish-momentum results, latest snapshot.

screening_get_bearish

Top bearish setup results, latest snapshot.

screening_get_supply_demand

Top supply/demand breakout results, latest snapshot.

screening_get_all

Latest snapshot across all three screens.

screening_get_by_criteria

Bullish results filtered by arbitrary criteria.

screening_run_screens

Recompute one screen (or all three) and persist it (mutates).

portfolio_add_position

Add/average into a position (mutates).

portfolio_get_my_portfolio

Full portfolio snapshot with live P&L.

portfolio_remove_position

Remove shares from a position (mutates).

portfolio_clear_portfolio

Remove every position; requires confirm=True (mutates).

portfolio_risk_adjusted_analysis

ATR-based position sizing/stop/target.

portfolio_compare_tickers

Side-by-side ticker comparison (auto-uses your portfolio).

portfolio_correlation_analysis

Correlation matrix and diversification metrics.

portfolio_get_risk_dashboard

Total value, sector exposure, and risk metrics.

portfolio_check_position_risk

Pre-trade risk check for a hypothetical trade.

portfolio_get_regime_adjusted_sizing

Position size scaled by detected market regime.

portfolio_get_risk_alerts

Current sector/position/portfolio risk alerts.

portfolio_watchlist_create

Create a named watchlist (mutates).

portfolio_watchlist_add

Add a ticker to a watchlist (mutates).

portfolio_watchlist_remove

Remove a ticker from a watchlist (mutates).

portfolio_watchlist_brief

Intelligence brief for every symbol on a watchlist.

portfolio_journal_add_trade

Log a new open trade (mutates).

portfolio_journal_close_trade

Close an open trade; PnL computed automatically (mutates).

portfolio_journal_list_trades

List journal trades, optionally filtered.

portfolio_journal_review

Full detail for a single journal trade.

portfolio_get_strategy_performance

Strategy performance analytics, with optional comparison.

backtesting_run_backtest

Run a single-strategy backtest: metrics, trades, analysis.

backtesting_optimize_strategy

Grid-search a strategy's parameters.

backtesting_walk_forward_analysis

Rolling optimize/test windows to gauge robustness.

backtesting_monte_carlo_simulation

Bootstrap-resample trades for a return/drawdown distribution.

backtesting_compare_strategies

Backtest multiple strategies on the same symbol and rank them.

backtesting_list_strategies

List every rule-based strategy template with default parameters.

backtesting_backtest_portfolio

Backtest one strategy across multiple symbols.

backtesting_parse_strategy

Parse a natural-language description into a strategy + parameters (BYOK LLM).

backtesting_run_ml_strategy_backtest

Backtest an ML-enhanced strategy (adaptive, ensemble, regime-aware).

backtesting_train_ml_predictor

Train a random-forest ML predictor for trading signals.

backtesting_analyze_market_regimes

Detect bear/sideways/bull regimes for a symbol.

backtesting_create_strategy_ensemble

Backtest a weighted ensemble of base strategies.

research_run_comprehensive

Comprehensive web-search-backed research on a financial topic.

research_analyze_company

Comprehensive research on a specific company.

research_analyze_sentiment

Market sentiment analysis for a topic or sector.

Setup

Requires Python 3.12+ and uv. Core install: uv sync --extra dev (or pip install "maverick-mcp-server"). Optional extras: --extra backtesting, --extra research. Copy .env.example to .env and configure (optional). Start: make dev (HTTP) or make dev-stdio (stdio). Add the JSON config to Claude Desktop.

claude_desktop_config.json

{
  "mcpServers": {
    "maverick-mcp": {
      "command": "uvx",
      "args": [
        "--from",
        "maverick-mcp-server",
        "maverick-mcp",
        "--transport",
        "stdio"
      ]
    }
  }
}

Generate and validate client config

Fit and risk

Best for

Individual traders and investors who want local, self-hosted stock analysis with Claude Desktop.
Users seeking an all-in-one tool combining technical indicators, screening, and portfolio tracking.
Developers and data enthusiasts who appreciate FastMCP and modern Python tooling (uv).

Not for

Users needing institutional-grade data or professional backtesting features.
Those expecting a pre-seeded stock universe (screening requires fetching price history first).
Users wanting a hosted, zero-configuration solution (self-hosting requires technical setup).

Required permissions

Runs locally and listens on a port (e.g., 8003) for HTTP transport.
Accesses the filesystem to read config and database (SQLite file).
May access Redis and external network (yfinance for market data).

Risks and side effects

Investment risk: market data may be delayed or inaccurate; past performance does not guarantee future results.
Relies on third-party data sources (yfinance) and optional external services (Exa, LLM APIs) that may be interrupted or change.
Self-hosting configuration can be error-prone, especially the Windows Claude Desktop cwd issue.
The built-in screening and backtesting tools are for educational purposes only, not investment advice.

Troubleshooting

Ensure the HTTP endpoint has a trailing slash: http://localhost:8003/mcp/ ; otherwise tool registration fails.
If research tools are unavailable, check that the research extra is installed and LLM_PROVIDER, LLM_API_KEY, LLM_MODEL, and EXA_API_KEY are set.
If backtesting tools are unavailable, install the backtesting extra.
If screening results are empty, fetch price history for relevant tickers (market_data_get_price_history) before running screens.
Windows users experiencing ModuleNotFoundError should wrap the command in cmd.exe to force the cwd.

Use cases

Perform technical analysis on stocks using natural language, e.g., "Show me the RSI and MACD analysis for NVDA".

Run bullish, bearish, and supply/demand screens to spot trading opportunities.

Manage personal portfolio: add positions, view live P&L, get risk dashboard.

Backtest and optimize strategies using VectorBT.

Conduct deep research and sentiment analysis on companies with LangGraph and research tools.

Supported clients

Claude DesktopFull support

CursorFull support

Claude CodeFull support

WindsurfFull support

Related MCP servers

Compare with other MCP servers