Repolode
Submit
HKUDS

HKUDS/Vibe-Trading

An open-source personal trading agent that runs research, backtests, paper trading and broker connections from one command, via CLI, Web UI, API or MCP.

View on GitHubWebsite
Stars
37,781
Forks
6,122
Open issues
56
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?

Vibe-Trading is a personal trading agent that gives an AI agent a set of trading capabilities. It covers market research, strategy generation, backtesting, and paper trading, and it can connect to a range of brokers and data sources. The project is installed with pip as vibe-trading-ai and can be used from a command-line interface, a Web UI, a REST API, or as an MCP server for other agents.

The agent works on quantitative tasks such as writing and testing strategies, computing risk metrics like Sharpe, Sortino, drawdown and Monte Carlo results, and producing research reports. Recent releases add a paper-trading tool that matches orders against completed daily bars without reaching a live broker, saved paper accounts, scheduled reports delivered by email or IM channels, and token usage tracking.

The project covers multiple markets, including A-shares, Hong Kong, US equities, futures, crypto and forex, and supports several LLM providers. It is MIT licensed and written in Python with a FastAPI backend and a React frontend.

Key features

  • 01Installs with one pip command and runs from CLI, Web UI, REST API or MCP server.
  • 02Runs backtests and reports metrics such as Sharpe, Sortino, drawdown and Monte Carlo results.
  • 03Provides a paper-trading tool and saved paper accounts that never reach a live broker.
  • 04Connects to multiple brokers and data sources across A-shares, Hong Kong, US, futures, crypto and forex.
  • 05Produces research reports and can send scheduled reports by email or IM channels.

Use cases

  • Testing a quantitative trading idea with historical data before risking capital.
  • Running daily paper-trading experiments against completed bars.
  • Giving an existing AI agent access to backtesting and market data through MCP.
  • Producing scheduled research summaries for a portfolio.

Quick start

  1. 1

    Install the package from PyPI

    pip install vibe-trading-ai
  2. 2

    Open the CLI, Web UI, API or MCP entry point described in the README

Things to know

  • Some broker integrations are incomplete, and Robinhood options execution remains unavailable.
  • Live options trading is blocked, and the project's changelog notes ongoing unreleased fixes.

In collections

Open-source AI agent frameworks & appsMCP servers

Topics

#ai-agent#algorithmic-trading#backtesting#quantitative-finance#llm#mcp#python#fintech#multi-agent#trading

FAQ

What is Vibe-Trading?
An open-source personal trading agent that runs research, backtests, paper trading and broker connections from one command, via CLI, Web UI, API or MCP. Vibe-Trading is a personal trading agent that gives an AI agent a set of trading capabilities. It covers market research, strategy generation, backtesting, and paper trading, and it can connect to a range of brokers and data sources. The project is installed with pip as vibe-trading-ai and can be used from a command-line interface, a Web UI, a REST API, or as an MCP server for other agents.
Is Vibe-Trading free to use? What license does it have?
Vibe-Trading is open source under the MIT license. Check the license terms before using it in commercial products.
Is Vibe-Trading still maintained?
Actively maintained. The last commit to Vibe-Trading was on Nov 6, 2026.
What can I use Vibe-Trading for?
Testing a quantitative trading idea with historical data before risking capital. Running daily paper-trading experiments against completed bars. Giving an existing AI agent access to backtesting and market data through MCP. Producing scheduled research summaries for a portfolio.
How do I install Vibe-Trading?
Install the package from PyPI: pip install vibe-trading-ai Open the CLI, Web UI, API or MCP entry point described in the README
Where can I get Vibe-Trading?
The source code and installation instructions are on GitHub: https://github.com/HKUDS/Vibe-Trading

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

Related projects