Browse / 🤖 AI & Agents

Vibe-Trading preview

HKUDS

Vibe-Trading

"Vibe-Trading: Your Personal Trading Agent"

Stars
38.5k
Language
Python
Category
🤖 AI & Agents
ai-agentalgorithmic-tradingbacktestingfintechllmmcpmulti-agentpythonquantitative-financetrading

English | 中文 | 日本語 | 한국어 | العربية | Español | Bahasa Indonesia

Vibe-Trading Logo

Vibe-Trading: Your Personal Trading Agent

One Command to Empower Your Agent with Comprehensive Trading Capabilities

📄 Read the Technical Report  arXiv: 2611.24845

Star History: GitHub Trending Trendshift: daily ranking Star History: global star ranking

GitHub topic rankings by stars · Nov 4, 2026
#1 Backtesting  ·  #3 Algorithmic Trading  ·  #4 Quantitative Finance

PyPI total downloads README views

Python FastAPI React PyPI License
Feishu WeChat Discord

Website  ·  Docs  ·  News  ·  Features  ·  Shadow Account  ·  Demo  ·  Quick Start  ·  Examples  ·  API / MCP  ·  Roadmap  ·  Contributors

pip install vibe-trading-ai

📰 News

  • 2026-11-19 📄 Technical report published — The report describes how Vibe-Trading traces results to source data and runs, preserves research revisions, and enforces user-authorized limits on live trading. Vibe-Trading: Towards Agent-Native Trading

  • 2026-11-18 📈 Taiwan equity backtests and execution fixes — TWSE (.TW) and TPEx (.TWO) stocks now support daily-bar backtests via Yahoo Finance. The Korea, Taiwan and Vietnam backtest engines use the simulated fill price, including extra slippage, to check daily price limits. Intraday backtest requests for these markets are rejected before fetching data. #2174 · #2175 Switching between listed sessions reuses the sidebar list. The bundled AKShare skill supports current and legacy endpoint names. Regression tests cover attachment filenames, configuration lookup order, and MCP market-data previews and artifact parts. On main, for the next pip release.

  • 2026-11-17 📊 Portfolio risk details and reusable options analysis — Risk X-Ray adds observed drawdown periods and signed per-holding volatility contributions. Options Lab exports calculated snapshots as JSON and scenario grids as CSV, and diagnoses observed IV slices with explicit coverage limits; the diagnostics do not certify an arbitrage-free surface. Agent/MCP tools compare named option strategies on shared price and IV grids. Fix malformed update metadata and failed upgrade verification, direct-mode model discovery, attachment filenames, missing-price handling and stale session-list responses. On main, for the next pip release.

  • 2026-11-16 🛠️ More reliable reports, memory search and messaging — Shadow Account reports keep unmeasured P&L unavailable instead of showing zero, including cached HTML/PDF exports. Memory search repairs legacy indexes; partially failed PDF OCR is marked degraded. Fix Signal pairing, preserve Napcat text when image metadata is malformed, narrow Anthropic temperature retries to the relevant validation error, and support current and legacy AKShare history endpoints. See CHANGELOG. On main, for the next pip release.

Earlier news
  • 2026-11-15 📡 Unified market-data gateway and ready-to-use CSVs — Historical bars and US option snapshots share a local data gateway with source fallback, shared caching and bounded concurrency. Responses preserve source, currency and price-adjustment metadata, with explicit coverage and freshness information. Agent market-data reads deliver CSVs alongside JSON. For supported requests naming instruments and one explicit YYYY-MM-DD range, the runtime constrains dates before acquisition; calculation warmup is requested separately, and existing indicator/backtest lookbacks remain available. On main, for the next pip release.

  • 2026-11-14 🔌 Unified LLM gateway and version checks — Model providers now share one catalog and configuration layer across Agent, CLI, API and Swarm. Client settings stay isolated, native adapters respect explicit proxy settings, and cross-origin redirects are blocked before credentials or request bodies are forwarded. Check for new versions from Web settings. Desktop gains support for signed updates with verification and recovery; distribution requires a configured update feed and publisher certificates. On main, for the next pip release.

  • 2026-11-13 🛠️ Research delivery and calculation tools — Research answers stream without automatic numeric redaction, forced rewrites, or format-based completion gates. Interrupted committee runs preserve partial results and can resume unfinished roles while reusing completed work. Full market-data references feed shell-free QuantLib calculations, and committee roles share source evidence. Options Lab and options_payoff calculate stock-plus-options strategies and compare multiple alternatives in one call, returning explicit costs, payoff bounds, and comparison directions. Effective backtest fees and truncation/error states remain explicit; live-trading authorization limits remain in place. Unreleased.

  • 2026-11-12 🛠️ Backtest timing, report figures and research completion fixes — Shadow Account now uses the requested starting capital; token-truncated worker replies no longer count as completed. Committee PMs can synthesize valid upstream reports without fetching the same data again, and the native Anthropic adapter defaults to a configurable 16,384 output tokens. Generated strategies reject detected extra shifts on output targets unless explicitly declared intentional, while indicator lookbacks remain valid. Grounding now accepts supported magnitude spellings and precise observed declarations without weakening evidence checks. SPY/GLD backtests match an independent ledger on all 206/140 fills and total returns. Unreleased.

  • 2026-11-11 🧾 Paper-trade partial fills and ledger reconciliation — Opt into per-order volume caps, partial fills and reported order states for A-share 5-minute paper simulations (#2128). Independently reconcile fees, cash, positions and NAV arithmetic from returned fills with the new paper_trade_reconcile Agent/MCP tool (#2197). Correctly handle supported wrapped percentage headers and explicitly quoted document integers while preserving their source labels (#2185, #2192). Also fix strict factor benchmark inputs (#2195) and prevent timed-out channel turns from returning an earlier answer (#2194). Thanks @shadowinlife, @he-yufeng, @Jackzigen and @divya0795! On main, for the next pip release.

  • 2026-11-10 📊 Interactive IV surfaces and options analysis fixes — Edit strike/expiry IV grids in Options Lab, with observed/interpolated labels and no extrapolation (#2186). Fix low-price presets, quantities, expiry dates, retries, language-switch loading and the 20-leg limit. Bare three-digit Zhengzhou contracts such as CF501 now use China futures rules (#2193). Thanks @he-yufeng and @GPT-AGI! On main, for the next pip release.

  • 2026-11-09 🧭 Export a selected revision comparison; backtests refuse codes they cannot place — Download the exact pair of research revisions you are comparing as JSON or Markdown from the Web, API or CLI (#2189). A backtest code no market rule recognises (such as local:BHP.AX or 7203.T) is now refused before any data is fetched instead of running on A-share rules; bare six-digit and .SS Shanghai codes still work. OpenRouter replies no longer show the model name twice (#2191), and an empty revision id in research-artifact export is now an error. Thanks @he-yufeng and @Gloubiboulgator! On main, for the next pip release.

  • 2026-11-08 📝 Research revision comparison and full-history export — Compare content, evidence and metadata between any two revisions (#2168), and export the complete research history as JSON or Markdown (#2170). Switching research records now ignores stale responses, and Markdown exports preserve original content in isolated code blocks. Also fix missing factor weights, partial-return attribution, insufficient IC samples, non-finite valuation inputs and margin calculations for negative entry prices. On main, for the next pip release.

  • 2026-11-07 🧪 Research Lab: trading costs and out-of-sample testing — Rerun saved price inputs under different fees and slippage; select parameters using training results only, then freeze them for out-of-sample tests and a separate final holdout. The Research Lab shares Web, CLI, API and Agent/MCP workflows, with all eleven UI languages. Also correct window-boundary returns and factor warmup/long-window calculations, and add 20 offline task-replay cases that keep missing evidence visible. Versioned research artifacts can now store structured financial facts (#2164), and bare Chinese futures contract codes resolve correctly (#2177). On main, for the next pip release.

  • 2026-11-06 📈 Implied-volatility grids and paper-trading fixes — Build strike-by-expiry IV grids from supplied option observations, with observed/interpolated labels and no extrapolation (#2172). Extreme numeric inputs now produce finite results or explicit errors. A-share 5m market orders blocked by a price limit are canceled immediately (#2176); the matcher also processes symbols chronologically so later proceeds cannot fund earlier orders. On main, for the next pip release; public 5m requests still support one symbol per step.

  • 2026-11-05 🚀 v0.1.19 released (Release notes, pip install -U vibe-trading-ai): research that outlives the chat — save versioned research artifacts with their evidence and revise them, reuse backtests from earlier turns and sessions, and edit a research goal in place. New opt-ins: a five-minute A-share paper simulation, and Gildata financial queries, stock screening and Shannon portfolio diagnosis (QVeris is removed; an existing GILDATA_TOKEN alone now enables nothing). Also Argentina (BYMA) backtests, Parallel web search, Codex OAuth model discovery, and both 0.1.18 regressions fixed. 116 merged pull requests plus 21 landed through maintainer commits, from 26 contributors (CHANGELOG). Fixed today: a backtest reports the benchmark it actually evaluated, long/short books get a risk x-ray, GTJA alpha 17 stays missing while its inputs warm up, the Shannon panel speaks all eleven UI languages, and two reviewed contributor fixes that had not landed are in (#2042 TTM sums across a missing quarter, #1985 3:2 split detection). Thanks @rudycelekli, @he-yufeng, @Jackzigen, @Shizoqua, @cgycorey, @dajiaohuang, @zeus229, @alexwan02, @BHhhh-123, @remote-controlled-man, @shadowinlife, @TWJxi, @lishuanghong1989-netizen, @TerryF7, @davidalmeida90, @DrDLyuan, @goatyyc, @hnq-ai, @kadyrbekovhamit-cyber, @MetaAviator, @octo-patch, @RowanAldean, @VectorPeak, @wavebyrd and @yufengsika! On main after the release (next pip release): backtest benchmarks are bought at the first evaluated open, as the strategy is, instead of from the last warm-up close; excess return no longer includes that move (AAPL 20/60 SMA example: 22.5% → 18.4%).

  • 2026-11-04 📊 Gildata financial queries, stock screening and Shannon portfolio diagnosis — Screen stocks by market cap and retrieve their half-year revenue in one request, or inspect portfolio factor exposures and each holding’s contribution to volatility. All nine Gildata features are opt-in, including for existing token holders. Free data sources never fall back to Gildata. #2117 · #2118 · #2119 Also clarified next-bar backtest fills, fixed false Swarm timeouts on Unicode heartbeats and sped up research artifact deduplication. #2137 · #2155 · #2159

  • 2026-11-03 🛠️ Runtime integrity and localization follow-ups: Preserve U+2028/U+2029 inside trace, audit-ledger and session-search JSONL records, keep ledger appends valid after such payloads, normalize underscored model labels, and preserve handoff summaries during layer-2 compaction. Also accept BOM-prefixed frontmatter and keep the welcome surface aligned in RTL locales. Add regression coverage.

  • 2026-11-02 🛠️ Frontmatter, result safety and welcome polish: Unquote quoted scalar metadata, fall back to a readable truncation when an oversized result cannot be encoded, and refine the welcome layout and featured cards. Add regression coverage.

  • 2026-11-01 🛠️ Session timestamps, theme contrast and welcome interactions: Stamp session updates in UTC, keep dark-mode primary buttons legible, and make side-panel examples clickable with clear category labels. Add regression coverage.

  • 2026-10-31 🛠️ Welcome and ticker parsing polish: Move the welcome surface to a warm paper palette and two-column layout, make featured examples and side-panel actions clearer, and stop trailing ticker letters from being interpreted as A-share or A/H constraints.

Full news archive · Changelog

✨ Key Features

🔍 Market Research: Explore prices, filings and ETF holdings; Value companies with DCF and peer multiples; Track earnings and schedule a premarket brief.🐝 Research Teams: 30 preset teams for equities, quant, crypto and risk; Follow the bull case, bear case and risk review; Assign research tasks in Pixel Office.
📊 Quant Lab: Benchmark 478 alphas across five libraries; Backtest across markets and test out of sample; Model options payoffs, Greeks and portfolio risk.👥 Trading & Portfolio: 19 broker connectors with per-broker permissions; Track paper fills or set limits for supported live trades; Replay your own trading rules with Shadow Account.

💡 What Is Vibe-Trading?

Vibe-Trading turns finance questions into research reports, runnable backtests and paper portfolios. You can also authorize a supported broker to execute live trades within a mandate you set and can halt. Your funds stay with the broker.

✨ What You Can Do

Task Where to start
Research markets, filings or uploaded documents Ask in Web chat or attach a report.
Backtest a strategy or benchmark alphas Inspect charts and trades in Run Detail, or compare factors in /alpha-zoo.
Simulate trading or replay a journal Track daily fills and NAV in /paper-trading. Use /journal and /shadow in the CLI to review your trades.
Combine accounts or compare allocations Open /portfolio for holdings and proposed allocation changes. Broker support
Explore options and correlations Open /options for multi-leg payoffs and Greeks, or /correlation for market regimes.
Coordinate or repeat research Run a Swarm team, assign work in Pixel Office or schedule a brief in /scheduled.
Revisit research with dated evidence Search /research-artifacts, compare any two revisions, and export a full series or selected comparison as JSON or Markdown. Versioned research

⚡ Quick Example

Once you have run vibe-trading init, you can ask for a backtest in plain English or benchmark an existing alpha library from the terminal.

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 20

Open Run Detail to compare the crossover with buy-and-hold, inspect the trades behind a drawdown and check the data source on the run card.

Portfolio drawdown periods

In Run Detail > Portfolio Studio > Risk X-Ray, inspect the five deepest observed drawdowns, with peak, trough, recovery and duration. A drawdown still open at the end of the input window is reported separately even when it is outside the deepest five. The same drawdown.episodes data is saved in artifacts/risk_xray.json, summarized in artifacts/risk_xray.md, and returned by the portfolio_risk_xray Agent/MCP tool.

This is a constant-weight return model, not account NAV, strategy P&L or a forecast. Backtests use the average executed basket weights; the model itself does not include fees or cash flows. Duration counts observed bar intervals from peak to recovery, or to the window's end if unrecovered, not calendar days. Daily, weekly and monthly observations retain their original labels.

A missing interior return or non-positive/non-finite model wealth makes the periods unavailable, rather than an empty set or zero risk. Values within a relative 1e-12 of a running peak are treated as at that peak to avoid an extra recovery bar caused only by floating-point noise; smaller differences are not distinguished. The JSON report declares this as recovery_rtol. Existing scalar statistics and older reports without episode metadata remain supported.

🧪 Research Workflow

Step In the crossover example
Plan Identify the trading pair, date range, moving averages and benchmark from the prompt.
Ground Fetch historical bars for BTC-USDT. Check the source and dates before interpreting the result.
Execute Write the strategy and run the backtest. The run directory holds the strategy, trades and equity curve.
Validate Compare the result with buy-and-hold. Inspect turnover, drawdown and validation warnings alongside return.
Revisit Open the saved run or ask a follow-up with different parameters. You can also save a dated research conclusion and revise it when new evidence arrives.

📡 Data Sources & Smart Fallback

The loader supports 28 market-data sources. With source: "auto", it infers the market from the symbol and tries the configured sources in order.

Market How a request is filled
A-shares and Hong Kong Start with Tencent. If it cannot supply the bars, try Eastmoney and then the remaining sources for that market.
US equities Start with Yahoo. Stooq and Sina are the next two sources in the chain.
Crypto Try OKX, Binance and CCXT before falling back to other sources.
Futures Chinese contracts can use AKShare or local bars. Supply local bars for global contracts.

You can also choose a source explicitly. For example, installing vibe-trading-ai[ashare] adds the token-free BaoStock loader for daily A-share bars. The data source reference lists the full chains and credential requirements.

🔩 Detailed Capabilities

Market research

Feature What you can do
Filings and holdings Compare a manager’s SEC 13F holdings across quarters, or look through an ETF to its underlying positions. Each holdings report carries its reporting date.
Company valuation Use /dcf to see how a valuation changes with the discount rate and terminal growth. Use /comps for peer multiples, or build a linked three-statement model and export the results to Excel.
Papers and documents Search arXiv and OpenAlex for a signal, inspect the authors’ definition, then test it on your own universe. Upload a filing, spreadsheet or slide deck for questions about a specific exhibit.
Stock screening With Gildata enabled, screen companies by market cap and retrieve half-year revenue in one request. Screening and portfolio factor diagnosis each have their own opt-in.

Quant research

Feature What you can do
Alpha Zoo Benchmark signals from Qlib, Alpha101, GTJA and the academic and fundamental libraries. Compare information coefficients across horizons, then check whether the result survives a walk-forward test.
Backtest engines Model A-share price limits, crypto funding or futures margin. Use a shared capital pool for a cross-market strategy and inspect the individual fills behind its returns.
Options Lab Build a spread leg by leg. Compare its expiry payoff with the Greeks and see how the position responds to a change in spot price and implied volatility. Export the exact analysis request and result as JSON or the full expiry and scenario grid as CSV. The options_strategy_compare Agent/MCP tool compares up to four strategies on a shared spot/IV grid with baseline-relative P&L.
Portfolio risk Inspect drawdowns, estimate value at risk and expected shortfall, or use Brinson attribution to separate allocation from stock selection. Risk X-Ray reports each holding's signed contribution to portfolio volatility, including hedges that offset risk. The correlation view shows when markets start moving together.
Strategy tracking Register a strategy with its evidence and track how its Sharpe ratio or factor information coefficient changes as new results come in.

Risk X-Ray includes risk_contributions in the Agent tool response and saved JSON/Markdown reports. Components use weight × Cov(asset return, portfolio return) / portfolio volatility, with the same sample and annualization as total volatility. Each assets row names its symbol; risk_contribution is annualized decimal volatility and contribution_pct is in percentage points. They add up to total volatility, not portfolio return. A hedge can have a negative risk share, and another holding can exceed 100%; shares are unavailable for zero volatility or an insufficient sample. This is a historical weighted-basket diagnosis, not realized PnL attribution or a forecast.

Accounts and trading

Feature What you can do
Portfolio Bring read-only broker holdings into one snapshot or import an extraETF export. Enter target weights to preview the value change for each holding.
Paper trading Advance a saved account one trading day and inspect fills, cash and NAV. You can also replay an A-share session in five-minute steps in a separate simulation. The paper_trade_reconcile agent/MCP tool independently checks fees, cash, positions and NAV arithmetic without market or broker access.
Shadow Account Import your broker’s trade journal, inspect fees and trading habits, then compare the actual trades with a rule-based replay.
Broker connections Set up a connection in the CLI or Web and test it with a read. For supported live profiles, set symbol, order and exposure limits before enabling trading. The kill switch halts live activity.

Teams and recurring research

Feature What you can do
Swarm and Pixel Office Run a bull–bear investment committee or a quant research desk and inspect the workers’ reports. In Pixel Office, assign a task to a teammate and record the result on the shared board.
Scheduled briefs Start from a premarket brief, an earnings tracker or a portfolio checkup. Choose a schedule and send the report to a configured channel such as Slack, Telegram or email.
Research history Search older sessions and reopen a saved run. Save a dated conclusion with its evidence and compare revisions when the investment case changes.
Run records Inspect the data source, model, execution record and metric references behind a backtest. Reopen the saved trades and equity curve to check a reported figure.

Broker support at a glance

Compare read-only, paper and live support by broker

Read means a dedicated read-only profile. Sim means paper orders through a broker sandbox or local simulation. Live means live orders under a mandate.

Yes indicates a declared capability. A dash means no matching profile is declared.

Broker Read Sim Live
alpaca Yes Yes Yes
binance Yes Yes Yes
dhan Yes Yes —
etoro Yes Yes Yes
futu Yes Yes Yes
ibkr Yes — —
kis Yes Yes —
longbridge Yes Yes —
mstock Yes — —
mt5 Yes Yes Yes
okx Yes Yes Yes
robinhood Yes — Yes
scalable Yes — —
shoonya Yes Yes —
tiger Yes Yes Yes
toss Yes — —
trading212 Yes — —
upbit Yes Yes —
zerodha Yes Yes —

Full permissions by profile

🎬 Demo

Web · From prompt to backtestCLI · Interactive terminal
demo-web-research.mp4
cli-playbooks.mp4
Send a strategy request, follow the agent’s work and inspect the resulting run.Browse the research playbooks and open the pre-market brief in the terminal.

🚀 Quick Start

Terminal. Install the package with Python 3.11+, choose a model provider in the setup wizard, then open the interactive CLI.

pip install vibe-trading-ai
vibe-trading init
vibe-trading

Web app. Docker serves the frontend and API together at http://localhost:8899. From a fresh checkout, configure your provider before starting the container.

git clone https://github.com/HKUDS/Vibe-Trading.git
cd Vibe-Trading
cp agent/.env.example agent/.env
# Set your provider and API key in agent/.env.
docker compose up --build

The PyPI package includes the CLI, API and MCP server. For a frontend development setup or a local model, see the installation guide.

🖥 CLI Reference

Start vibe-trading for the interactive terminal. Type /help at the prompt to browse commands or use these shortcuts during research.

Command Use it to
/model Switch the provider or model for the session.
/history Browse and resume a prior conversation.
/show <run-id> Inspect the results of a saved run.
/pine Export the current strategy as a Pine Script indicator.

From the shell, vibe-trading alpha list browses the factor libraries and vibe-trading playbook list shows the scheduled research templates.

💡 Examples

  • Compare stocks. Ask “Compare AAPL and MSFT over 2024. Which month accounts for the largest gap in returns?” Resume the conversation later with /history.
  • Review your trades. Use /journal to analyze your exported trade history, then /shadow to compare your trades with a rule-based alternative.
  • Run an investment committee. Choose a stock and follow the bull and bear cases through to the final report.
vibe-trading --swarm-run investment_committee '{"target":"TSLA","market":"US"}'

🌐 API Server

Start a local server with vibe-trading serve --port 8899. In a second terminal, run curl http://localhost:8899/runs to list saved runs. Use the returned ID with GET /runs/{run_id} to inspect its results.

To run a research session over HTTP Endpoint
Create the conversation POST /sessions
Send the research question POST /sessions/{id}/messages
Follow progress as it happens GET /sessions/{id}/events

Open http://localhost:8899/docs to try the endpoints interactively. Authenticated clients send the API key in the Authorization: Bearer header.

🔌 MCP Plugin

MCP clients can launch the same package over stdio. Add this entry to Claude Desktop’s claude_desktop_config.json, restart the client and call list_skills.

{
  "mcpServers": {
    "vibe-trading": { "command": "vibe-trading-mcp" }
  }
}

For other clients or a custom output directory, see the MCP configurations.

🏛 Ecosystem

Vibe-Trading is part of the HKUDS agent ecosystem:

NanoBot
Ultra-Lightweight Personal AI Assistant
AI-Trader
Agent-Native Signal & Copy Trading Platform
CLI-Anything
Making All Software Agent-Native
OpenSpace
Self-Evolving AI Agent Skills
ClawTeam
Agent Swarm Intelligence

🗺 Roadmap

Area Available now Next
Paper trading Daily accounts persist cash, fills and NAV between sessions. A separate simulation replays an A-share session in five-minute steps. Opt-in volume-share matching (#2128, draft).
Research Run cards record metrics and evidence for a completed backtest. Dated research artifacts keep revision history and are searchable. Structured financial facts in research artifacts (#2164, in review).
Options Lab Explore multi-leg payoffs, Greeks and spot × IV scenarios; edit IV grids and check observed slices for model-price inconsistencies at /options. These checks do not calibrate or certify an arbitrage-free surface.

Observed-slice diagnostics use your spot price and continuously compounded risk-free rate and dividend yield to reconstruct European call model prices. They check vertical spr

README truncated — open on GitHub for the full file.

More in 🤖 AI & Agents

32.6k

decolua

9router

Unlimited FREE AI coding. Connect Claude Code, Codex, Cursor, Cline, Copilot, Antigravity to FREE Claude/GPT/Gemini via 40+ providers. Auto-fallback, RTK -40% tokens, never hit limits.

Indiesiders