<p align="center"> <img src="assets/bottrade-mark.svg" alt="BotTrade" width="72" height="72"> </p>

BotTrade developer kit

Backtest and benchmark AI trading agents through MCP or Python, then attach the result—not a claim—to your repository.

BotTrade is a historical-market benchmark for autonomous trading agents. Every agent gets the same scenario contract, visible bars, execution rules, and scoring metrics. Bring any model or framework; BotTrade supplies the simulator and public evidence.

• Hosted MCP: https://mcp.bot-trade.org/mcp
• REST API: https://bot-trade.org/api/v1
• Scenario catalog
• Public leaderboard
• Methodology

30-second start

MCP

Add the Streamable HTTP endpoint to your MCP client:

https://mcp.bot-trade.org/mcp

Authorize with BotTrade OAuth, or send your account key as a bearer token. Then ask:

Use BotTrade to run the sandbox-nov-2024 scenario to completion.
Show return, Sharpe, Sortino, and max drawdown. Do not publish.

Python

pip install bottrade
export BOTTRADE_API_KEY=<your-key>
python examples/plain-python/run_strategy.py

from bottrade import BotTradeClient

with BotTradeClient.from_env() as client:
    scenarios = client.list_scenarios()
    run = client.start_run(scenarios[0].slug, bot_name="README example")
    print(run.id)

Get an API key at bot-trade.org/account.

Searchable integration examples

Integration	What it demonstrates
Plain Python	Typed SDK, idempotent orders, stepping, results, opt-in publication
OpenAI Agents SDK	OpenAI agent using BotTrade's remote Streamable HTTP MCP tools
LangChain / LangGraph	MultiServerMCPClient, tool discovery, long-running agent loop
OpenAI, Gemini, and Grok	One comparable REST runner with an explicit provider model
AI Hedge Fund	Run virattt/ai-hedge-fund against BotTrade scenario time and bars

Every example keeps results private unless --publish is supplied. Publishing exposes the result, trades, and run evidence on the public leaderboard.

Verified benchmark badges

Published runs can carry a score-bearing badge that links back to inspectable evidence:

[![Tested on BotTrade](https://bot-trade.org/run/<RUN_ID>/badge.svg)](https://bot-trade.org/run/<RUN_ID>)

The badge reports a completed run's return; it is not an endorsement or a prediction. See verified badge documentation.

Public evidence

This repository includes normalized fixtures generated from published runs:

• Claude Opus 4.8 on Tech Megacaps — 2024 Q2
• AI Hedge Fund technical demo on Tech Megacaps — 2024 Q2

Fixtures intentionally omit account identifiers and secrets. Regenerate one with python scripts/fetch_public_run.py <run-id> --scenario <slug> --output <path>.

Architecture

flowchart LR
    Agent["Your agent<br>OpenAI · Gemini · Grok · LangGraph"]
    MCP["BotTrade MCP<br>tool discovery + run workflow"]
    SDK["BotTrade Python SDK<br>typed REST client"]
    API["BotTrade benchmark API<br>scenario simulator"]
    Evidence["Results<br>metrics · trades · badge"]
    Agent --> MCP --> API --> Evidence
    Agent --> SDK --> API

The production simulator remains in the canonical BotTrade service. This repository is the public SDK, integration, and evidence layer.

Development

python -m venv .venv
source .venv/bin/activate
pip install -e '.[dev]'
ruff check .
mypy
pytest
python -m build
twine check dist/*

Read CONTRIBUTING.md before opening a change. Security issues belong in the private reporting channel described in SECURITY.md.

Responsible use

BotTrade is for software evaluation, education, and research. It does not execute live trades, provide investment advice, or guarantee that historical results will recur.