Maverick MCP Server

Financial data analysis and portfolio tracking for your AI assistant.

Data last scanned yesterday · Reviewed 3 weeks ago

3 1 dangerous tool52 tools★ 643PythonMIT

Overview

MaverickMCP is a Python-based server designed for individual traders to integrate financial market data directly into their AI interface. It provides over 30 tools for stock screening, technical analysis, and portfolio monitoring without requiring API keys for core functions, as it leverages the yfinance library for market data.

Our verdict

This server carries a high risk profile with a security score of 33/100, primarily due to the lack of any authentication method, meaning anyone with access to your configuration can trigger these tools. Furthermore, it lacks a dedicated read-only mode and is maintained by the community rather than an official financial vendor. It is suitable for local, personal use by traders aware of these access risks but should not be deployed in environments requiring strict security controls.

No API key required for core market data tools.
Zero direct dependencies simplifies the supply-chain surface.
Over 30 tools for technical analysis and portfolio tracking.
Fast installation process using modern uv dependency management.

No authentication layer allows unauthorized access to config users.
Absence of read-only mode for safety.
Community-maintained project lacking formal security audits.

Tools

Tool	Description	Risk
market_data_get_price_history	OHLCV price history for a ticker, smart-cached.	low
market_data_get_price_history_batch	Price history for multiple tickers at once.	low
market_data_get_quote	A single quote, TTL-cached.	low
market_data_get_stock_fundamentals	Valuation, financials, and trading stats.	low
market_data_get_market_overview	Indices, sector performance, top movers, and volatility.	low
market_data_get_chart_links	Static external chart links for a ticker.	low
market_data_clear_market_cache	Clear cached quotes (mutates cache state).	medium
technical_get_rsi_analysis	RSI reading and signal label.	low
technical_get_macd_analysis	MACD reading, signal label, and crossover state.	low
technical_get_support_resistance	Support/resistance levels.	low

Compatibility

Client	Local	Docker	Remote	Read-only
ChatGPT
Claude Desktop
Cursor
VS Code
Windsurf

Frequently asked questions

›Do I need an API key to access market data?

No, the core tools for market data, technical analysis, and portfolio tracking rely on yfinance and do not require any API keys.

›What are the optional extras I can install?

You can install optional extras for backtesting (powered by VectorBT) and deep research (powered by LangGraph and Exa) by including [backtesting] or [research] during installation.

›What are the system prerequisites for running this server?

The server requires Python 3.12+ and the uv package manager. While SQLite is the default for persistence, Redis is supported for enhanced caching, and PostgreSQL can be used for data storage.

›Does the research extra require additional configuration?

Yes, using the [research] extra requires you to bring your own LLM key and an Exa API key.

›How does the server handle data caching?

The server uses a tiered caching system that checks memory, then Redis or SQLite, with a graceful fallback if Redis is not currently running.

›How can I connect this to Claude Desktop?

You can connect via stdio by adding the server configuration to your Claude Desktop MCP settings, using either the uvx command for a published package or running the server module from a local directory.