Skip to content
Get packs

Guide · Open source

How to use Vibe-Trading: install with pip and run your first agent

Updated · written with AI from the project's own documentation and checked against our catalogue data

Vibe-Trading is an open-source Python project from HKUDS that gives an AI agent trading-related capabilities such as research and backtesting. This guide walks you through checking the requirements, installing it from PyPI, and understanding what it is and is not designed to do. By the end you will know how to get it running and where to look next in the official docs.

Vibe-TradingHKUDS/Vibe-Trading · ★ 37.4k · MIT

Quick start

  1. 1Check Python. Confirm you have Python 3.11 or newer installed.
  2. 2Install from PyPI. Run pip install vibe-trading-ai in your terminal.
  3. 3Read the API and MCP section. Find out how to connect Vibe-Trading to your own agent or tools.
  4. 4Test with backtesting. Run research and backtests before using any broker connection.
  5. 5Try paper trading. Use simulated orders to check behaviour without real money.
  6. 6Review the docs. Use the official docs at vibetrading.wiki for configuration details.

What is Vibe-Trading and who is it for?

Vibe-Trading describes itself as a personal trading agent. Its README says it is installed with one pip command and then connected to your own tools. Its topics include backtesting, multi-agent systems, quantitative finance and MCP.

It suits developers who want an AI agent for quantitative trading research, people who want to test strategies with backtesting before risking money, students learning quantitative finance, and users who already run an LLM-based agent and want trading tools through the API or MCP.

  • Good fit: quant research, strategy backtesting, learning quantitative finance, adding trading tools to an existing agent.
  • Poor fit: anyone wanting a system that makes profitable trades. The README makes no performance claims and calls this research software.
  • Poor fit: anyone needing a polished no-code app, since the README centers on command-line installation and Python setup.

What do you need before you install Vibe-Trading?

The README lists Python 3.11 or newer as the prerequisite. You also need pip, the Python package installer, to install from PyPI. The README does not list other system requirements in the portion we reviewed, so check the official docs for anything beyond this.

  • Python 3.11+
  • pip (installed with most Python distributions)
python --version

Install Vibe-Trading from PyPI

The README's install method is a single pip command that pulls the package from PyPI under the name vibe-trading-ai. Run it in your terminal or in a virtual environment.

pip install vibe-trading-ai

Connect Vibe-Trading to your agent, API or MCP setup

The README's navigation includes an API / MCP section, and its description says you connect the agent to your own tools after installing. The README excerpt we reviewed does not show the exact connection steps, so follow the API / MCP section of the README and the official docs at https://vibetrading.wiki/docs/ for configuration details.

  • Use the API server if your own agent calls trading tools over HTTP.
  • Use MCP if your LLM client supports the Model Context Protocol.

Understand paper trading and broker connections

The README mentions paper trading and portfolio connections to brokers, including IBKR and Futu. Recent news entries in the README describe saving IBKR and Futu gateway addresses through the Web or CLI, and Robinhood read-only portfolio reads. Treat broker connections as something to set up carefully and test first.

Paper trading means simulated orders with no real money. Use it to check behaviour before connecting anything that touches a live account.

  • Start with backtesting and paper trading, not live orders.
  • Do not connect a wallet or buy tokens. The README warns that related token and memecoin claims are not official.

Common pitfalls and how to avoid them

The README contains a security warning: the X account VibeTrading_HKU, a Virtuals project, and a token contract are not official Vibe-Trading assets. The project says it has never launched or endorsed any token or memecoin.

The README's news shows many changes labelled as unreleased and on the main branch for the next release. If you need stability, check the release status before relying on a specific behaviour.

  • Only trust the official website and repository for downloads and links.
  • Do not sign anything or connect a wallet because of a third-party token promotion.
  • Expect unreleased changes on main; pin a version you have tested.

What people use it for

  • Backtesting a trading idea before committing money.
  • Running quantitative finance research as a learning exercise.
  • Giving an existing LLM agent access to trading tools through the API or MCP.
  • Testing portfolio read-only connections to brokers such as IBKR or Futu.

Frequently asked questions

How do I install Vibe-Trading?

Install it from PyPI with pip install vibe-trading-ai. The README lists Python 3.11 or newer as the prerequisite.

What are the requirements for Vibe-Trading?

The README lists Python 3.11+. Other system requirements are not covered in the excerpt we reviewed, so check the official docs.

Is Vibe-Trading free?

The project is released under the MIT licence, as shown in the README badge. Check the LICENSE file in the repository for the full terms.

Does Vibe-Trading make profitable trades automatically?

No. The README does not claim performance and describes the project as research software. It is not a ready-made system that makes profitable trades.

What are alternatives to Vibe-Trading?

Projects in the same space include FinRL, Qlib, Freqtrade and Backtrader. Compare them on your needs, such as backtesting, agent integration, or live trading support.

Is Vibe-Trading connected to a crypto token?

No. The README states that the X account, Virtuals project and token contract associated with it are not official, and the project has never launched or endorsed any token or memecoin.

New agent skills, repos and AI tools — every week

The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.

Our verdict

Vibe-Trading suits Python developers, quant students and people already running an LLM agent who want backtesting and trading tools. It is not a no-code app or a system that promises profits, so people wanting a turnkey trading bot should look elsewhere.

Install options & downloads