HKUDS / Vibe-Trading

2026-04-01 · Source: Github Trending: All languages · Field: Finance & Economics — Capital Markets & Investment Management, FinTech & Digital Financial Services · Depth: Advanced, extended

Summary

Vibe-Trading is an open-source research workspace designed to convert natural-language finance questions into actionable analysis. It integrates market-data loaders, strategy generation, backtest engines, reporting, and persistent research memory. The platform also supports autonomous trading via authorized brokers like Robinhood Agentic Trading, operating strictly within user-defined limits and offering an instant kill switch. The recent v0.1.9 release on 2026-06-01 introduced six new broker connectors (Tiger, Longbridge, Alpaca, OKX, Binance, Futu), enabling paper-account order placement and mandate-gated live trading for five of them. Key features include a "Shadow Account" for analyzing personal trading behavior from broker journals, a "Research Goal" runtime for structured task management, and an "Alpha Zoo" offering 452 pre-built quant alphas for benchmarking. The project has also seen extensive security hardening and UI/CLI enhancements.

Key takeaway

For AI Engineers and Data Scientists developing or evaluating trading strategies, Vibe-Trading provides a robust, open-source platform to accelerate your workflow. You can use its natural-language processing to quickly prototype strategies, backtest across diverse markets, and analyze your own trading behavior with the Shadow Account. Integrate its 452 pre-built alphas for rapid benchmarking, or utilize its multi-agent swarm capabilities for collaborative research, ensuring your autonomous trading agents operate within defined safety mandates.

Key insights

Vibe-Trading empowers users to transform natural-language finance queries into executable trading analysis and bounded autonomous trading.

Principles

• Agentic trading requires strict user mandates and kill switches.
• Comprehensive financial analysis benefits from diverse data sources.
• Quant alpha research needs rigorous benchmarking and lookahead guards.

Method

Vibe-Trading's research workflow involves planning, grounding with market data/documents, executing strategy code or analysis tools, validating outputs with metrics, and delivering reports or exports.

In practice

• Use `vibe-trading alpha bench` to test pre-built alphas.
• Analyze personal trading biases with the Shadow Account feature.
• Export strategies to TradingView, TDX, or MetaTrader 5.

Topics

• Algorithmic Trading
• AI Agents
• Quantitative Finance
• Backtesting
• Broker Integrations
• Alpha Research

Code references

• HKUDS/Vibe-Trading
• HKUDS/Vibe-Trading
• HKUDS/nanobot
• HKUDS/AI-Trader
• HKUDS/CLI-Anything

Best for: AI Engineer, Machine Learning Engineer, Data Scientist

Related on AIssential

• Building a Hyperliquid AI Agent Trader From Scratch — An AI agent can automate high-frequency perpetual trading on platforms like Hyperliquid by integrating account setup, API access, and a research-to-execution pipeline.
• 10 GitHub Repositories to Master Quant Trading — Quant trading is a systematic, data-driven process requiring comprehensive tools for strategy, risk, and execution, not isolated ideas.
• TauricResearch / TradingAgents — TradingAgents uses specialized LLM-powered agents to simulate complex financial trading decisions through collaborative analysis and debate.
• Market-Research Firm AlphaSense Clinches $7.5 Billion Valuation in New Funding Round — AI-driven market intelligence platforms are attracting substantial investment, with valuations rapidly increasing.
• 40 AI Agents. One Live Trading Market. No Hidden Results. — Transparency in AI trading through live, multi-agent competition builds trust and accountability.
• hummingbot / hummingbot — Hummingbot democratizes high-frequency trading via an open-source framework for automated strategies across diverse exchanges.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Github Trending: All languages.