dr-manhattan

Expose a CCXT-style unified API for prediction market trading.

196|34|Updated Oct 25, 2025
One-click install
npx skills add https://github.com/guzus/dr-manhattan --skill dr-manhattan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dr-manhattan
Source: https://github.com/guzus/dr-manhattan/tree/main
Command: npx skills add https://github.com/guzus/dr-manhattan --skill dr-manhattan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires py-clob-client, python-dotenv, and includes scripts (resource) components.

What problem does it solve?

Dr. Manhattan provides a single, CCXT-style API to interact with multiple prediction market platforms, simplifying integration and automation.

Core Features & Use Cases

  • Unified interface exposes Polymarket, Kalshi, Opinion, Limitless and Predict.fun through a single API.
  • Market discovery, price fetching, order placement, and balance/position tracking across exchanges.
  • MCP server integration enables Claude/AI agents to trade via the model context protocol.
  • Real-world use: build cross-exchange bots that fetch markets, compare prices, and execute strategies.

Quick Start

Install dependencies and start using the library to fetch markets and place orders across supported exchanges. Example: initialize Polymarket, Opinion, Limitless, and PredictFun objects and call fetch_markets.

Frequently Asked Questions about dr-manhattan

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I use a CCXT-style API for prediction markets?

You can use a unified CCXT-style API to browse, search, and trade prediction markets across Polymarket, Kalshi, Opinion, Limitless, and Predict.fun. It simplifies integration by exposing market discovery, price fetching, and order placement through a single interface.

Can I run cross-exchange trading bots for prediction markets?

Yes, you can build cross-exchange bots that fetch markets, compare prices across platforms, and execute automated trading strategies. The unified API supports market-making, order management, and position tracking across multiple prediction market exchanges.

How does MCP server integration work with prediction market trading?

MCP server integration enables Claude and AI agents to trade on prediction markets via the model context protocol. This allows AI agents to interact with exchanges, fetch market data, and manage orders programmatically through the unified API.

Do I need trading credentials to fetch prediction market data?

No, trading credentials are optional. You can browse markets, search for specific events, and fetch prices without authentication. Credentials are only required when placing orders or checking balances and positions.

What Python version is required for prediction market API integration?

Python 3.11 or higher is required to use the unified prediction market API. The implementation depends on py-clob-client and python-dotenv, and requires network access for API calls to supported exchanges.

What prediction market platforms does the unified API support?

The unified API supports Polymarket, Kalshi, Opinion, Limitless, and Predict.fun. You can initialize individual exchange objects and call methods like fetch_markets to interact with each platform through a consistent interface.