Agentโ™ฅ๏ธŽAge
Catalog

io.github.hmesutozsoy/oddsrail

Officialdeprecated

by hmesutozsoy ยท Python

Prediction markets for AI agents: Polymarket + Kalshi data, signals, attributed order routing

oddsrail (io.github.hmesutozsoy/oddsrail) MCP Server

This Model Context Protocol (MCP) server provides prediction-market access for any MCP-compatible agent. It integrates Polymarket and Kalshi data to support market search, orderbooks, price history, and positions, plus order routing. On Polymarket, it includes on-chain builder-code attribution.

๐Ÿ› ๏ธ Key Features

  • Prediction-market access across Polymarket and Kalshi
  • Market search, orderbooks, price history, and positions
  • Order routing
  • On Polymarket: on-chain builder-code attribution (project builder code registered at 0 bp)
  • Two premium signal tools: in-play overshoot/fade detection, and resolution dispute-risk

๐Ÿš€ Use Cases

  • Trading or analysis workflows for prediction markets via AI agents
  • Market data retrieval and state inspection (orderbooks, price history, positions)
  • Automated routing of orders to Polymarket/Kalshi

โšก Developer Benefits

  • Works with Claude Code, Claude Desktop, and any MCP-compatible client
  • Provides structured market search and trading-related market views

โš ๏ธ Limitations

  • Described scope focuses on Polymarket and Kalshi data and trading-related capabilities only

Topics

ai-agentskalshimcpmodel-context-protocolpolymarketprediction-marketstrading