Polymarket Agents

Build and deploy autonomous AI agents to trade on the Polymarket prediction market using an open-source Python framework.

Build and deploy autonomous AI agents to trade on the Polymarket prediction market using an open-source Python framework.

The gist

Polymarket Agents is an open-source developer framework created by Polymarket. It provides a set of Python utilities for building and deploying autonomous AI agents that can trade on the Polymarket prediction market. The framework is designed to help developers connect to market data, incorporate external information sources like news and web search, and execute trades programmatically, standardizing the process for creating data-driven trading bots.

What it does

  • Build autonomous AI agents for trading on prediction markets.
  • Integrate with the Polymarket API to fetch market data and execute trades.
  • Source data from news providers, web search, and other services to inform agent decisions.
  • Incorporate Retrieval-Augmented Generation (RAG) using local or remote vector databases.
  • Interact with the framework and manage agents through a command-line interface.
  • Utilize comprehensive LLM tools for prompt engineering.

How it works

As a self-hosted Python framework, users clone the GitHub repository and configure a local environment with a funded Polygon wallet and an OpenAI API key. Developers interact with the tool via a command-line interface to query market data, run agents, and execute trades. The framework connects to the Polymarket API, processes information using LLMs, and signs orders programmatically. It is open-source and free to use under the MIT License.

Best for

This framework is best for Python developers who want to build and deploy custom, automated AI trading agents for the Polymarket prediction market.

Watch out for

The Polymarket Terms of Service explicitly prohibit US persons and individuals from certain other restricted jurisdictions from trading on the platform, including via the API and agents built with this framework.