MCP Predictive Market
Aggregates prediction market data from 5 major platforms (Manifold, Polymarket, Metaculus, PredictIt, Kalshi), enabling users to search markets, compare odds across platforms, detect arbitrage opportunities, and track predictions through natural language.
README
<p align="center"> <h1 align="center">MCP Predictive Market</h1> <p align="center"> <strong>Query and analyze prediction markets through Claude</strong> </p> <p align="center"> <a href="https://github.com/EricGrill/mcp-predictive-market/blob/master/LICENSE"><img src="https://img.shields.io/badge/license-MIT-blue.svg" alt="MIT License"></a> <img src="https://img.shields.io/badge/tools-8-green.svg" alt="8 Tools"> <img src="https://img.shields.io/badge/platforms-5-purple.svg" alt="5 Platforms"> <img src="https://img.shields.io/badge/python-%3E%3D3.11-orange.svg" alt="Python >= 3.11"> <img src="https://img.shields.io/badge/tests-134%20passing-brightgreen.svg" alt="134 Tests"> </p> <p align="center"> <a href="#-quick-start">Quick Start</a> | <a href="#-available-tools">Tools</a> | <a href="#-supported-platforms">Platforms</a> | <a href="#%EF%B8%8F-configuration">Configuration</a> | <a href="https://github.com/EricGrill/agents-skills-plugins">Plugin Marketplace</a> </p> </p>
What is this?
An MCP (Model Context Protocol) server that aggregates prediction market data from 5 major platforms. Search markets, compare odds, detect arbitrage opportunities, and track predictions through natural language.
Works with Claude Desktop, Claude Code, Cursor, and any MCP-compatible client.
Part of the Claude Code Plugin Marketplace.
Quick Start
# Clone and install
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
Add to your Claude config and start querying markets:
"Find prediction markets about AI regulation"
Why Use This?
| Feature | Description |
|---|---|
| Multi-Platform Search | Query 5 prediction markets simultaneously |
| Arbitrage Detection | Find price discrepancies across platforms |
| Market Tracking | Build watchlists and monitor odds changes |
| Platform Comparison | Side-by-side odds for similar questions |
| Unified Data Model | Consistent market schema across all platforms |
Available Tools
Search & Discovery
| Tool | Description |
|---|---|
search_markets |
Search markets across all platforms by keyword |
list_categories |
Get available market categories |
browse_category |
Browse markets in a specific category |
Market Data
| Tool | Description |
|---|---|
get_market_odds |
Get current odds for a specific market |
compare_platforms |
Side-by-side comparison of similar markets |
Tracking
| Tool | Description |
|---|---|
track_market |
Add a market to your watchlist |
get_tracked_markets |
View all tracked markets with current prices |
Analysis
| Tool | Description |
|---|---|
find_arbitrage |
Detect price discrepancies between platforms |
Supported Platforms
| Platform | URL | Specialization |
|---|---|---|
| Manifold Markets | manifold.markets | Play money, wide variety |
| Polymarket | polymarket.com | Crypto, high liquidity |
| Metaculus | metaculus.com | Science, long-term forecasts |
| PredictIt | predictit.org | US politics |
| Kalshi | kalshi.com | CFTC-regulated, real money |
Configuration
Claude Desktop Setup
Add to your Claude Desktop config:
| Platform | Config Path |
|---|---|
| macOS | ~/Library/Application Support/Claude/claude_desktop_config.json |
| Windows | %APPDATA%\Claude\claude_desktop_config.json |
| Linux | ~/.config/Claude/claude_desktop_config.json |
{
"mcpServers": {
"prediction-market": {
"command": "uv",
"args": ["run", "--directory", "/path/to/mcp-predictive-market", "python", "-m", "mcp_predictive_market.server"]
}
}
}
Claude Code Setup
Add to your project .mcp.json or ~/.config/claude-code/mcp.json:
{
"mcpServers": {
"prediction-market": {
"command": "uv",
"args": ["run", "--directory", "/path/to/mcp-predictive-market", "python", "-m", "mcp_predictive_market.server"]
}
}
}
Examples
<details> <summary><b>Search & Discovery</b></summary>
"Find prediction markets about AI"
"What categories of markets are available?"
"Show me crypto markets on Polymarket"
"Browse politics markets"
</details>
<details> <summary><b>Market Analysis</b></summary>
"Get current odds for Manifold market abc123"
"Compare odds for 'Will Bitcoin hit $100k?' across all platforms"
"Show me the probability of a 2024 recession on different platforms"
</details>
<details> <summary><b>Arbitrage Detection</b></summary>
"Find arbitrage opportunities with at least 10% spread"
"Are there any markets with significantly different odds across platforms?"
"Show me the biggest price discrepancies right now"
</details>
<details> <summary><b>Market Tracking</b></summary>
"Track the Polymarket election market"
"Show all my tracked markets"
"What are the current prices on my watchlist?"
</details>
Development
# Clone
git clone https://github.com/EricGrill/mcp-predictive-market.git
cd mcp-predictive-market
# Install with dev dependencies
uv sync --extra dev
# Run tests
uv run pytest -v
# Run specific test file
uv run pytest tests/test_integration.py -v
Project Structure
src/mcp_predictive_market/
├── server.py # MCP server entry point
├── tools.py # Tool handler implementations
├── schema.py # Unified market data models
├── errors.py # Custom exceptions
├── rate_limiter.py # Per-platform rate limiting
├── adapters/ # Platform-specific adapters
│ ├── base.py # Adapter protocol
│ ├── manifold.py
│ ├── polymarket.py
│ ├── metaculus.py
│ ├── predictit.py
│ └── kalshi.py
├── analysis/ # Market analysis modules
│ ├── matching.py # Cross-platform market matching
│ └── arbitrage.py # Arbitrage detection
└── state/ # State management
└── memvid_client.py
Troubleshooting
<details> <summary><b>No results from a platform</b></summary>
- Platform API may be rate-limited - wait and retry
- Check platform is online: visit the website directly
- Some platforms filter certain market types
</details>
<details> <summary><b>Arbitrage opportunities not found</b></summary>
- Lower the
min_spreadparameter (default is 5%) - Try broader search terms
- Fewer opportunities exist in efficient markets
</details>
<details> <summary><b>Market not found</b></summary>
- Verify the market ID format (varies by platform)
- Ensure the market hasn't been resolved/closed
- Check you're using the correct platform name
</details>
Related Projects
- Claude Code Plugin Marketplace - Discover more MCP plugins
- MCP Proxmox Admin - Manage Proxmox VE through Claude
- MCP Memvid State Service - Persistent state for MCP servers
Contributing
Contributions welcome!
- Fork the repository
- Create feature branch:
git checkout -b feature/my-feature - Make changes and test:
uv run pytest - Commit:
git commit -m 'Add my feature' - Push:
git push origin feature/my-feature - Open a Pull Request
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.