Vibe-Trading: A personal trading agent that turns finance questions into research reports, runnable backtests and paper portfolios.
What problem it solves
Finance research often requires combining market data, source documents, valuation methods, strategy testing and portfolio analysis across multiple workflows. This project provides an agent-based interface that turns those questions into research reports, backtests and paper portfolios, with optional supported live broker execution under user-defined limits.
Key features
Natural-language research and backtesting — Ask for a market or strategy analysis in plain English and receive a research report or runnable backtest. Runs can expose their strategy, trades, equity curve, benchmark comparisons and validation warnings.
Source: README
Multi-source market data — The loader supports 28 market-data sources and can infer a source chain from the requested market, or accept an explicitly selected source.
Source: README
Quant and options analysis — Benchmark alpha libraries, run cross-market backtests, build multi-leg options positions and inspect Greeks, payoff behavior and portfolio risk.
Source: README
Paper trading and Shadow Account — Simulate trading one trading day at a time, replay A-share sessions in five-minute steps and compare a broker journal with a rule-based replay.
Source: README
Versioned research and coordinated teams — Save dated conclusions with evidence, compare revisions and export research history. Swarm teams and Pixel Office can divide research work among bull, bear, quant or risk roles.
Source: README
Quick start
Copied verbatim from the project's README.
vibe-trading run -p "Backtest a BTC-USDT 20/50 moving-average crossover for 2024. Compare it with buy-and-hold and report maximum drawdown."
vibe-trading alpha bench --zoo gtja191 --universe csi300 --period 2018-2025 --top 20Who it may suit
This may suit individual traders, quantitative researchers and finance professionals who want agent-assisted research, backtesting and paper-portfolio workflows. It may also suit developers building brokerage, MCP or API integrations around finance tooling.
Topics to discuss
Which research workflow should the agent handle most reliably when evidence is incomplete or conflicting?
How should live-trading mandates, permission boundaries and the kill switch be represented across Web, CLI, API and Agent/MCP interfaces?
Which additions to the versioned research artifact format would improve auditability, reproducibility and comparison across revisions?
Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.
Read the original README https://github.com/HKUDS/Vibe-Trading#readme
ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.
Comments
…