Discover and explore top open-source AI tools and projects—updated daily.
agent-nextAI agent-powered paper trading simulator for decentralized prediction markets
Top 95.9% on SourcePulse
This project provides a paper trading simulator for Polymarket, specifically designed for AI agents. It allows users to test trading strategies on real Polymarket order books with simulated funds, offering a zero-risk environment to track performance, backtest strategies, and compete on a leaderboard. The tool is beneficial for AI researchers and traders looking to validate strategies against live market conditions without financial risk.
How It Works
The simulator meticulously replicates Polymarket's exchange mechanics to ensure realistic trading conditions. It features level-by-level order book execution, accurately consuming liquidity at each price level. The exact fee model and slippage tracking are implemented, allowing paper trading P&L to closely mirror potential real-world P&L within the spread. It supports complex multi-outcome markets beyond simple binary YES/NO and includes a limit order state machine with Good-Til-Cancelled (GTC) and Good-Til-Date (GTD) capabilities.
Quick Start & Requirements
pip install polymarket-paper-trader), ClawHub (npx clawhub install polymarket-paper-trader), or from source (uv pip install -e ".[dev]").examples/ directory.Highlighted Details
Maintenance & Community
This project is part of agent-next, an open research lab for autonomous agents. Specific community links (e.g., Discord, Slack) are not detailed in the README.
Licensing & Compatibility
The project is released under the MIT license, which is permissive for commercial use and integration into closed-source projects.
Limitations & Caveats
This is a paper trading simulator and does not involve real money. It specifically simulates the Polymarket exchange and its associated mechanics. Live API integration tests require network connectivity.
1 month ago
Inactive