Open-source · MIT · Python
The AI agent core
for prediction markets.
polybot is a production-grade trading platform with a native MCP server — letting Claude, GPT, and other agents analyse markets, size positions, and execute trades across Polymarket, Kalshi, Opinion, and Binance. Not a no-code workflow. A domain agent.
MCP server built-in
25+ typed tools for market analysis, position management, strategy assessment, and approval-gated execution. Your agent gets a trading desk.
One engine, four venues
Polymarket CLOB, Kalshi (CFTC-regulated), Opinion DEX, and Binance for hedging. Unified risk and position model across all of them.
Shadow mode first
Every strategy runs paper-trading by default. Promote to live only when the metrics prove out. Human-in-the-loop approvals on agent-driven trades.
Ten strategies, not one
Pluggable BaseStrategy interface. Enable, shadow, or disable each one from the CLI.
Arbitrage
Low riskBuy YES + NO when combined < $1. Near-risk-free if sized right.
Statistical arbitrage
Medium riskTrade correlated markets that diverge from historical spread.
AI model
Medium riskLet an LLM or ML model predict true probability, trade the gap.
Spread farming
Low riskPassive market making — earn the bid-ask spread.
Copy trading
Medium riskMirror whale wallets you have back-tested as profitable.
Resolution arbitrage
Low riskMispricing near expiry when outcome is nearly certain.
Not n8n. An agent core.
n8n is a generic workflow runner. polybot is a domain agent with native types for markets, orders, positions, and risk — and an MCP server that exposes all of it to Claude, GPT, or any MCP-compatible client. The LLM isn't a node in a workflow; it's a first-class principal with tools, memory, and approval-gated actions.
Read the full comparison →$ polybot mcp start
MCP server listening on stdio
# From Claude Desktop / any MCP client:
→ list_markets(category="politics", limit=10)
→ analyze_strategy("arbitrage", window_days=30)
→ suggest_strategy_improvements("spread_farming")
→ place_order(market_id=..., side="YES", size=50)
[approval required — shadow: paper-filled]Install in four commands
pip install polybot-trader
cp .env.example .env # add POLYMARKET_PRIVATE_KEY, POLYMARKET_PROXY_ADDRESS
polybot db init
polybot strategy enable arbitrage
polybot strategy shadow arbitrage --enable
polybot startDashboard comes up at http://localhost:8000/ui. Strategies run in shadow mode until you explicitly promote them to live.
Need an agent system built like this?
Cryptuon builds production AI agents, MCP integrations, and trading systems. polybot is our open-source showcase.