Repolode
Submit
wshobson

wshobson/maverick-mcp

A local MCP server that gives AI assistants stock quotes, technical analysis, portfolio tracking and optional Python backtesting via Yahoo Finance.

Open-source alternative to:TradingView

View on GitHubWebsite
Stars
721
Forks
169
Open issues
7
Last commit
yesterday

Repolode describes open-source projects and does not give investment advice. Trading and investing can lose money: assess the risk yourself and follow the laws where you live before using this project.

What is it?

MaverickMCP is an MCP (Model Context Protocol) server for stock market analysis. It lets an AI assistant such as Claude Desktop, Cursor or VS Code call tools that fetch prices and fundamentals from Yahoo Finance through yfinance, compute indicators such as RSI and MACD, and show observed support and resistance levels. Core tools need no API key, and the server stores portfolios, watchlists and a trade journal in a local SQLite database by default.

The project also provides stock screening over symbols loaded into its database, portfolio risk checks, and a trade journal that records trades locally without sending orders to a broker. An optional backtesting extra adds strategy tests, parameter optimization, walk-forward analysis and machine learning strategies built on Python. A separate research extra adds web-based company and sector research with source citations, using a configured language model and Exa or SearXNG search.

The server can run over STDIO for a single client or over Streamable HTTP on localhost, and it ships with a Dockerfile. It is intended for educational and informational use and does not provide financial advice.

Key features

  • 01Fetches quotes, historical OHLCV prices and fundamentals from Yahoo Finance without an API key.
  • 02Computes RSI, MACD and observed support and resistance levels.
  • 03Tracks portfolio positions, P&L, watchlists and a local trade journal.
  • 04Runs bullish, bearish and supply/demand stock screens over loaded symbols.
  • 05Provides optional backtesting tools with strategy templates and ML models.
  • 06Offers optional web research with source citations via a configured LLM.

Use cases

  • Asking an AI assistant to summarize a stock's technical indicators and recent price range.
  • Keeping a local record of holdings and checking risk exposure before a trade.
  • Testing a trading idea on historical data with Python backtests.
  • Logging and reviewing past trades in a private local journal.

Quick start

  1. 1

    Clone the repository

    git clone https://github.com/wshobson/maverick-mcp.git
  2. 2

    Enter the project folder

    cd maverick-mcp
  3. 3

    Install the core dependencies

    uv sync
  4. 4

    Start the server for your MCP client (replace the path with your checkout's absolute path)

    uv run --directory /absolute/path/to/maverick-mcp maverick-mcp --transport stdio

Things to know

  • The Streamable HTTP server has no authentication, so keep it local unless you add access controls.
  • Market data comes from Yahoo Finance and can be delayed, incomplete, or wrong; the project does not give financial advice.
  • It requires Python 3.12 or later and the uv tool, and optional research can incur model and search provider charges.

In collections

MCP servers for Claude CodeClaude Code tools, skills & pluginsOpen-source AI agent frameworks & appsMCP servers

Topics

#stock-market#mcp#finance#technical-analysis#backtesting#portfolio-tracking#yfinance#claude#ai-agents#anthropic

FAQ

What is maverick-mcp?
A local MCP server that gives AI assistants stock quotes, technical analysis, portfolio tracking and optional Python backtesting via Yahoo Finance. MaverickMCP is an MCP (Model Context Protocol) server for stock market analysis. It lets an AI assistant such as Claude Desktop, Cursor or VS Code call tools that fetch prices and fundamentals from Yahoo Finance through yfinance, compute indicators such as RSI and MACD, and show observed support and resistance levels. Core tools need no API key, and the server stores portfolios, watchlists and a trade journal in a local SQLite database by default.
Is maverick-mcp free to use? What license does it have?
maverick-mcp is open source under the MIT license. Check the license terms before using it in commercial products.
Is maverick-mcp still maintained?
Actively maintained. The last commit to maverick-mcp was on Nov 6, 2026.
What can I use maverick-mcp for?
Asking an AI assistant to summarize a stock's technical indicators and recent price range. Keeping a local record of holdings and checking risk exposure before a trade. Testing a trading idea on historical data with Python backtests. Logging and reviewing past trades in a private local journal.
How do I install maverick-mcp?
Clone the repository: git clone https://github.com/wshobson/maverick-mcp.git Enter the project folder: cd maverick-mcp Install the core dependencies: uv sync Start the server for your MCP client (replace the path with your checkout's absolute path): uv run --directory /absolute/path/to/maverick-mcp maverick-mcp --transport stdio
Where can I get maverick-mcp?
The source code and installation instructions are on GitHub: https://github.com/wshobson/maverick-mcp

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects