Vibe-Trading
A free toolkit that lets you research and test stock or crypto trading strategies on historical data before risking real money, and optionally connect it to a real broker account with strict safety limits.
🔗 Visit Vibe-Trading
Description
Testing a trading idea properly means running it against years of historical price data first, not just trying it live and hoping — a step many beginner traders skip entirely. Vibe-Trading gives you the research and backtesting infrastructure to do this properly, plus an optional (and clearly labeled experimental) path to let an AI agent place real trades through a broker connection, gated behind explicit safety checks.
Vibe-Trading is a free, open-source (MIT license, ~27,400 GitHub stars) Python platform for quantitative trading research supporting multiple asset classes (stocks, crypto, forex), with backtesting, paper trading and live trading modes, a Robinhood broker connector, audit ledgers, and pre-trade advisory interfaces. The agentic autonomous trading feature is explicitly marked experimental and "use at your own risk," requiring mandatory user-committed trading mandates before any live capital is deployed, and the project includes security disclaimers warning against token/memecoin impersonation scams that sometimes target popular open-source trading tools.
💬 Our review
The short version: if you want a free, serious backtesting and research environment for trading strategies, Vibe-Trading is a substantial, well-adopted open-source option — but treat the "agentic live trading" feature with the caution its own documentation asks for.
The project is honest about its own risk profile, which is a genuinely good sign in a space full of overconfident "AI trading bot" projects — explicit safety gates, mandatory mandates before live trading, and scam warnings suggest the maintainers understand what can go wrong. That said, no backtest, however rigorous, guarantees future performance, and letting an autonomous agent place real trades introduces a whole new failure category beyond bad strategy (bugs, API misuse, unexpected market conditions) that a human trader wouldn't hit. For research and strategy development, it's a strong, free choice; for actual live automated trading with real money, go in with eyes open and start with the smallest possible amount you could afford to lose entirely.
💰 Pricing
📊 Global score
🤖 AI-enriched data
Entièrement gratuit et open source (licence MIT).
Pros
Gratuit, open source, très large communauté (27 400+ étoiles)
Recherche quantitative et backtesting rigoureux inclus
Fonctionnalités de sécurité explicites (mandats obligatoires, garde-fous)
Multi-actifs (actions, crypto, forex)
Cons
Trading autonome par agent IA explicitement marqué expérimental — risque réel avec de l'argent réel
Aucun backtest ne garantit une performance future
Nécessite des compétences techniques en Python pour un usage sérieux
Cible de scams potentiels (impersonation de tokens) selon les avertissements du projet lui-même