home / mcp / predictive market mcp server

Predictive Market MCP Server

Provides multi-platform prediction market data, search, tracking, and arbitrage analysis through MCP clients.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "ericgrill-mcp-predictive-market": {
      "command": "uv",
      "args": [
        "run",
        "--directory",
        "/path/to/mcp-predictive-market",
        "python",
        "-m",
        "mcp_predictive_market.server"
      ]
    }
  }
}

MCP Predictive Market lets you search, compare, and monitor prediction markets across multiple platforms through a single, unified interface. You can explore odds, spot arbitrage opportunities, and track markets in real time using Claude or any MCP-compatible client.

How to use

You connect your MCP client to the local or remote server and start issuing discovery and tracking commands. Use search to find markets by keyword, compare odds across platforms, and add markets to a watchlist to monitor changes over time. You can also request current odds for a specific market and look for arbitrage opportunities between platforms.

How to install

Prerequisites: You need Python installed and the UV toolchain available for running MCP servers.

Clone the project repository and install dependencies, then start the server.

git clone https://github.com/EricGrill/mcp-predictive-market.git
cd mcp-predictive-market
uv sync

# Run the server
uv run python -m mcp_predictive_market.server

Configuration

Configure the MCP server to run as a stdio (local) server that Claude can connect to. Use the following inline configuration snippet to register the server under the name prediction-market. This config uses UV as the launcher and starts the Python module that serves MCP endpoints.

{
  "mcpServers": {
    "prediction-market": {
      "command": "uv",
      "args": ["run", "--directory", "/path/to/mcp-predictive-market", "python", "-m", "mcp_predictive_market.server"]
    }
  }
}

Examples

Here are typical usages you can try after starting the server and configuring Claude or your MCP client.

Troubleshooting

If you don’t see results from a platform, wait a moment and retry as APIs may rate-limit or platforms may be temporarily offline. If arbitrage detection isn’t returning opportunities, try broader search terms or reduce the minimum spread threshold.

Development

For development workflows, run tests and use the provided entry points to verify integration with all supported platforms.

Available tools

search_markets

Search markets across all platforms by keyword.

list_categories

Retrieve available market categories.

browse_category

Browse markets within a specific category.

get_market_odds

Fetch current odds for a specific market.

compare_platforms

Compare odds for similar markets across platforms.

track_market

Add a market to your watchlist for monitoring.

get_tracked_markets

List all markets currently on your watchlist with current prices.

find_arbitrage

Identify price discrepancies between platforms.