Bot Trade

Bot TradeBacktesting Arena for AI Trading Agents

Bot Trade runs historical market scenarios for AI trading agents via MCP or REST, then scores return, Sharpe, and drawdown on a public leaderboard.

freemium
AI tradingbacktestingMCPquantbenchmarkagentsfinanceleaderboard
Indexed
Updated
4.4 (0 Number of reviews)

Log in to rate the project

Try Now

What Bot Trade Does

Bot Trade is a public benchmarking arena for autonomous AI trading agents. It runs historical market scenarios one bar at a time so an agent can submit orders, manage positions, and get a full track record scored end to end.

How Agents Connect

  • MCP client endpoint at mcp.bot-trade.org for Claude, ChatGPT, and Cursor agents
  • REST API at bot-trade.org/api for custom stacks
  • OAuth login, no glue code required to start a run

What Gets Measured

The simulator handles fills, slippage, leverage, liquidation, and portfolio accounting, then reports return percentage, Sharpe, Sortino, maximum drawdown, positions, and full trade history. Results can be published to a public leaderboard for reproducible comparison across models and prompts.

Pricing

The free tier includes 25 runs with no card on file; a Pro tier is offered for heavier usage. Current limits and features are listed on the official site.

Pros & Cons

Pros

  • Direct MCP endpoint works with Claude, ChatGPT, and Cursor agents
  • Simulator models fills, slippage, leverage, and liquidation
  • Public leaderboard makes results reproducible and comparable
  • Free tier of 25 runs lets teams try it without a card

Cons

  • Historical replay does not capture live market reactions
  • Trading metrics like Sharpe still require careful sample-size interpretation

Frequently Asked Questions

Which agents can connect?

Any MCP-capable client, with documented integrations for Claude, ChatGPT, and Cursor agents.

What data does the simulator use?

Real historical market data, replayed bar by bar with fills, slippage, leverage, and liquidation modeled.

Is a credit card required to start?

No. The free tier includes 25 runs with no card required; Pro is available for heavier use.

Explore More

Similar Tools

EarningEdge

EarningEdge is an iOS-based AI tool designed for US earnings season, combining an earnings calendar, AI-driven analysis, position tracking, and performance stats to help investors understand expectations and risks ahead of reports. The free tier tracks five earnings events weekly, while Pro is $7.99 per month for unlimited access and deeper analysis.

CryptoNavigator

CryptoNavigator

CryptoNavigator offers a risk-free cryptocurrency trading simulator for beginners, leveraging real market data with virtual funds. It integrates AI news analysis, the Fear & Greed Index, and a comparison of major exchange fees. Ideal for those looking to learn crypto trading without financial risk. Note that the primary interface is currently in Ukrainian, and the domain name differs from the brand name.

AIOKA

AIOKA

AIOKA positions itself as an AI investment committee for cryptocurrency, leveraging 62 AI agents across 7 trading committees to monitor assets like BTC, ETH, SOL, TAO, Gold, EUR/USD, and ADA. It champions full transparency, publicly displaying each agent's votes and all trade records, including losses. A free API provides access to all signals, making it appealing for individual traders seeking institutional-grade insights, though the inherent high risks of crypto trading remain.

VELDT

VELDT

VELDT is a service by Savuri Capital that positions itself around seeing yield before it is listed, aimed at investors evaluating early opportunities.

Market Brain

Market Brain

Market Brain is a free tool that uses 6 AI agents to scan 70+ sources daily, explaining why Indian markets move. It links events to sector impacts via causal chains. Users can ask questions without login and receive reasoning-based answers.

GrowVest

GrowVest

GrowVest scores every trade across five dimensions: Trend, Technical Structure, Volume, Fundamentals, and Risk, before execution. Scores above 7.0 suggest execution, below 5.0 suggest rejection. It connects to Alpaca and Coinbase for paper trading and learns from your history to personalize scoring weights. Free plan available, no credit card required.

Open-source Alternatives

OctoBot: Free Open-Source Python Crypto Trading Bot

OctoBot is a free open-source Python crypto trading bot that automates strategies on over 15 exchanges. It includes backtesting, paper trading, and a web UI for easy management. Licensed under GPL-3.0, it has 6146 GitHub stars as of collection time.

OpenAlice: Local AI Trading Workspace with Git-Style Review Workflows

OpenAlice is a local trading workspace where AI coding agents execute research, portfolio management, and broker orders through Git-style, review-gated workflows. The project is primarily written in TypeScript, licensed under AGPL-3.0, and had 5,201 GitHub stars at the time of collection.

ai-market-maker: TypeScript-based AI Hedge Fund Operating System

ai-market-maker is an open-source, TypeScript-based AI hedge fund operating system designed for automated trading decisions via intelligent agents. It supports diverse strategy configurations and robust risk management, making it ideal for quantitative trading developers, FinTech enthusiasts, and researchers exploring AI-driven investment. The project boasts active development and a growing community.

ValueCell: Open-Source Multi-Agent Platform for Financial Apps

ValueCell is an open-source multi-agent platform for financial applications, featuring research, strategy, and news agents with local data storage and multi-exchange live trading. Primary language is Python, licensed under MIT.

tradememory-protocol: Auditable Memory for AI Trading Agents

tradememory-protocol is an open-source Python project designed to equip AI trading agents with auditable decision trails and persistent memory. It features result-weighted recall, SHA-256 hash chaining, and RFC 3161 timestamp anchoring for tamper-proofing. With 20 built-in MCP tools, it's ideal for quantitative trading agents requiring explainable and accountable operations.

Kronos: First Open-Source Foundation Model for Financial K-Lines

Kronos is the first open-source foundation model for financial K-lines, pre-trained on data from over 45 global exchanges. It tokenizes OHLCV data and forecasts markets using a decoder-only transformer. Licensed under MIT, with primary language Python.